Co-localization of the elements calcium, potassium, sodium and magnesium with sequestering organelles has been achieved by application of two microscopy techniques on the same cell. Organelles were first localized by laser scanning confocal microscopy (LSCFM) using fluorescent organelle stains. The same cells were then analyzed for elemental distribution with ion microscopy. This approach has identified a perinuclear region of prominent total calcium concentration with the Golgi apparatus. Live cells were fluorescently stained with C6-NBD-cera-mide for labeling the Golgi apparatus prior to cryogenic preparation and freeze-drying, and imaged with LSCFM for Golgi localization; identical cells were then analyzed with ion microscopy to image subcellular distributions of total calcium, potassium, sodium and magnesium. In three cell lines, LLC-PKi porcine kidney epithelial cells, Swiss 3T3 mouse fibroblast cells and L5 rat myoblast cells, the Golgi regions contained significantly higher total calcium concentrations than any other region of the cell (as measured at the spatial resolution of ion microscopy of about 0.5 /an). Intracellular potassium, sodium and magnesium were homogeneously distributed throughout the cell and did not show this pattern. Measurements of depletion of calcium by exposure to calcium-free medium showed that the Golgi apparatus was substantially more resistant to calcium depletion than all other regions of these cells, but sequestered Ca2+ could be released from the Golgi by exposing the cells to calcium ionophore A23187. The Golgi apparatus appears to sequester about 5% of the total cell calcium in LLC-PKi cells, about 2.5% in 3T3 cells and L5 cells.

Intracellular calcium storage plays an essential role in the sequestration and release of calcium in physiological events (Berridge and Irvine, 1989; Ghosh et al. 1989; Meldolesi et al. 1990; Somlyo et al. 1985). The total calcium (both free and bound) in the cell is generally about ten thousand times larger than the ionized Ca2+. Although the endoplasmic reticulum (ER), mitochondria and cytoplasmic calcium-binding proteins are the most commonly identified Ca2+ storage sites, several lines of indirect evidence have also implicated the Golgi apparatus. An ATP-dependent Ca2+ transport capability has been demonstrated in isolated Golgi-rich membrane fractions from tissues such as rat liver (Hodson, 1978), intestine (Freedman et al. 1981) and mammary gland (Neville et al. 1981; West, 1981). In two studies of specialized secretory cells using an electron microprobe, a high-calcium perinuclear region has been observed and has been attributed to Golgi apparatus (Cameron et al. 1980; Roos, 1988); however, the Golgi cannot be clearly recognized in freeze-dried tissue cryosections because it consists of a stack of smooth membranes.

We have used laser scanning confocal fluorescence microscopy (LSCFM) (White et al. 1987) to localize Golgi by imaging C6-NBD-ceramide fluorescence in individual cells. We then employed ion microscopy (Castaing and Slodzian, 1962; Morrison and Slodzian, 1975) to image and localize intracellular total calcium in identical cells. This study provides direct evidence that the Golgi apparatus sequesters Ca2+, and that this structure is especially retentive of Ca2+ during depletion by exposure of cells to Ca2+-free extracellular medium. These observations suggest unique calcium sequestration and control associated with the Golgi apparatus.

Sample preparation requirements for correlated fluorescence and ion microscopy

The ion microscope, based on secondary ion mass spectrometry (SIMS), is a direct imaging mass spectrometer, which provides visual images of elemental (isotopic) gradients with a spatial resolution which can reach 0.5 /an (Morrison and Slodzian, 1975). The ion images reveal the distribution of the total (both free and bound forms) concentration of an element, and the images of several elements can be recorded from the same cell since the analysis is made by sputtering (eroding) the cell in the Z-direction. Due to the high vacuum requirements, live cells cannot be analyzed in the ion microscope. Instead, the specimen has to be freeze-fixed and freeze-dried in order to preserve the chemical and structural integrity of the living state (Chandra and Morrison, 1988). Recently, a freeze-fracture method was developed for ion microanalysis of cells in cultures (Chandra et al. 1986). In this method, cell cultures grown on a silicon substratum (a conducting substratum is a requirement for ion microscopy since the sample is held at 4500 volts) are flash-frozen and sandwich-fractured. Such a fracture produces a few randomly scattered areas where the fracture plane has split the apical side of the plasma membrane, thereby exposing the cell interiors for direct analysis. The fractured cells, therefore, consist of essentially the whole cells except for the outer leaflet of the apical membrane. Because the ions in the overlying extracellular medium are also removed along with the half-membrane, the subcellular elemental distributions can then be imaged from the fractured cells with the ion microscope. This method satisfies the requirements of preserving the native structural and chemical integrity (Ausserer et al. 1989; Chandra et al. 1987), and has been used for imaging ion transport after inhibition of the Na+-K+-ATPase of the plasma membrane with ouabain (Chandra and Morrison, 1985), and for subcellular elemental quantitation in cell cultures (Ausserer et al. 1989).

Since SIMS ion images reveal only intracellular elemental gradients, a correlative morphological technique is necessary for microfeature confirmation. The LSCFM is an ideal technique for this purpose since it is nondestructive and it can image the location of a particular organelle in 3-D. Although LSCFM is generally used for the study of live cells, the freeze-fractured, freeze-dried cells were used in this study so that the same cells could be studied by both fluorescence and ion microscopy. The freeze-dried cells did not pose any problems for fluorescence imaging with LSCFM and were comparable to live cells for Golgi localization.

Cell cultures, cryo sample preparation and fluorescence imaging

Three cell lines, LLC-PKj porcine kidney cells Swiss 3T3 mouse fibroblast cells and L5 rat myoblast cells (all from American Type Culture Collection), were seeded at a density of about 2x 105 cells per 100 mm in a Falcon plastic dish and incubated at 37 °C in a 5% CO2 atmosphere. Each Petri dish contained about 15 small wafer chips (approximately 1 cm2) of semiconductor grade N-type high-purity silicon (General Diode) in order to provide sister cells growing on the different chips to be subjected to separate desired treatments. After the cells reached about 80% confluency, they were stained with 5 μM C6-NBD-ceramide (using a bovine serum albumin complex of the fluorescent lipid N-[7-(4-nitrobenzo-2-oxa-l,3-diazole)]-6-aminocaproyl sphingosine) (Molecular Probe) for 10 min at 37 °C in serum-free medium and then washed with medium not containing the stain (Lipsky and Pagano, 1985a). The cells were then incubated for 45 min in the serum-free medium before the experiment. This incubation time was standardized to reveal a brightly labeled perinuclear Golgi apparatus with a faint cellular background resulting from other structures such as mitochondria and ER as described previously (Lipsky and Pagano, 1985a,b; Pagano and Sleight, 1985). The silicon chips containing sister cells were then divided into three treatments, (i) cells treated only with CQ-NBD cerimide, (ii) cells treated with Cg-NBD-cerimide and then exposed to Ca2+-free medium in the presence of 2 mM EGTA for 5 min, and (iii) cells treated with C6-NBD-cerimide and then exposed to 2 ííM A23187 in Ca2+-free medium in the presence of 2 mM EGTA for 5 min. 3T3 cells were treated with 1 μM A23187. In addition, LLC-PKj cells were exposed for 15 min to Ca2+-free medium prior to calcium imaging. After the desired treatments, the cells were flash-frozen in Freon-22 slush, freeze-fractured and freeze-dried (Chandra et al. 1986). After freeze-drying, the silicon chips containing cells were stored in a desiccator and kept in the dark to minimize dye bleaching. The individual silicon chips were then transferred to an air-tight sample chamber, which allowed optical measurements through a # 1.5, 22 mm2 Corning glass coverslip without rehydrating the cells. The fractured cells were easily recognized (Chandra et al. 1986), and Golgi apparatus was imaged in individual cells by recording the Cg-NBD-ceramide fluorescence with a Bio Rad LSCFM instrument (Wells et al. 1990). Following Golgi imaging, the identical cells were then analyzed with an ion microscope for imaging total calcium, potassium, sodium and magnesium distributions.

Volume measurements with LSCFM

To estimate the volume of the Golgi, 3-D images of the sample cells were acquired by recording a set of image planes separated by ΔZ≅0.5 pm. In each image, the area of the Golgi was calculated by measuring the area of the image where the brightness was above a user-defined threshold value. The Golgi volume of that section was found by multiplying the area by the section thickness of 0.5 /im. The sum of the volumes from each plane gives the total volume of the Golgi in the cell. These volumes were calculated for both live and freeze-dried cells. Live and freeze-dried cells were found to be comparable for Golgi localization and volume. Since both the cell nucleus and the outline of the cell could be recognized in dim Cg-NBD-cerimide fluorescence, the total cell and nuclear volumes were calculated by the same method.

Ion microscopy and calcium quantification

A CAMECA IMS-3f ion microscope, operating with a 5.5 keV mass-filtered primary ion beam of O2+ (the 100 nA beam current with a spot size of about 60 /on), was used to monitor positive secondary ions. The ion images were digitized directly from the microchannel plate/fluorescent screen detector assembly of the ion microscope using a charge-coupled device (CCD) imager (Photometries), digitized to 14 bits per pixel by a Photometries camera controller (Mantus and Morrison, 1990). The image processing was performed using a PDP 11/34 based system and a secondary ion mass image processing program (Ling et al. 1987). Ion images were calibrated using the cell matrix element ‘carbon’ as an internal standard (Ausserer et al. 1989). Seven successive images, carbon, potassium, sodium, calcium (three consecutive images) and magnesium were recorded. To obtain an accurate sampling of the calcium concentration in the Golgi, three consecutive calcium images were recorded (the individual calcium image is an integration of about a 0.3 pm thin slice of the cellular material). When high calcium intensities were found in all three images, the middle slice was selected for quantitative analysis. Following calcium depletion by ionophore treatment a region in approximate correspondence to the Golgi image was sampled for quantitative measurements. Three regions were selected for quantitative measurements in individual cells: a circular region covering most of the cell nucleus, a circular region covering most of the cell cytoplasm on one side of the cell and away from the Golgi apparatus and a small circular region covering only a portion of the Golgi apparatus.

The correlative fluorescence and ion microscopy approach reveals a perinuclear region of prominent total calcium concentration with the Golgi apparatus. Fig. 1 shows images of LLC-PK4 cells; in the left column the Golgi apparatus in several cells is mapped by LSCFM images of the fluorescence of C6-NBD-ceramide, and in the right column the corresponding SIMS images of total calcium distributions are shown from the same cells. The images in the top row show cells as grown and labeled. The middle row of images shows similarly labeled LLC-PKi cells that have been depleted of calcium after labeling by exposure to Ca2+-free medium for 5 min. The bottom row shows similarly labeled cells exposed to Ca2+-free medium for 5 min in the presence of 2 μM calcium ionophore A23187. The Cg-NBD-ceramide fluorescence image of LLC-PK4 cells (top left) has revealed the characteristic perinuclear brightly labeled Golgi apparatus clusters in each cell. The boundary of the individual cells can be recognized in dim fluorescence. Some cells have been numbered for the ease of matching identical cells between the fluorescence and the calcium images. Comparison with corresponding SIMS image of total calcium on the right show high calcium perinuclear patches that correlate well with the Golgi apparatus in each cell in the field of view. In each cell, the cell nucleus is distinctly lower in total calcium than the Golgi apparatus and noticably lower than the apparent level of calcium in the rest of the cell cytoplasm. Not resolved in these images of the regions of the cytoplasm are the ER and mitochondria. Thus the total calcium level here associated with ‘cytoplasm’ really represents an average over cytoplasm, mitochondria and ER. The cell labeled X was identified as injured, based on its K/Na ratio of 2 as compared to about 10 for other cells, and is disregarded in this analysis. Although the very edges of some cells also appear high in calcium, this effect cannot be distinguished from a preparation artifact due to adherence of the extracellular calcium from the nutrient medium and is also disregarded (Chandra et al. 1986). Unlike the preference for calcium sequestration in the Golgi apparatus and lower levels in the cell nucleus, potassium, sodium and magnesium were nearly homogeneously distributed throughout the cell (not shown). These expected uniform elemental distributions provide a control and necessary calibration for the SIMS images.

Fig. 1.

The LSCFM images of Cg-NBD-ceramide fluorescence (left column) revealing Golgi apparatus and corresponding ion microscopic total calcium images (right column) of fractured freeze-dried LLC-PKj cells. The cells in the top row were treated only with Cg-NBD-ceramide, the middle row represents cells treated with Cg-NBD-ceramide and then exposed to the Ca2+-free medium for 5 min in the presence of 2 mM EGTA, and the bottom row represents cells treated with Cg-NBD-ceramide and then exposed to 2 μM A23187 in Ca2+-free medium and 2 mM EGTA for 5 min. In ion microscope images, brightness indicates relative total calcium concentration within an image. The images of potassium, sodium and magnesium showed nearly homogeneous distribution throughout the cell (not shown). Calcium images were digitally processed and printed for optimum image quality. Ion image integration time for calcium images on the CCD imager was 100 s, Bar, 20 μm.

Fig. 1.

The LSCFM images of Cg-NBD-ceramide fluorescence (left column) revealing Golgi apparatus and corresponding ion microscopic total calcium images (right column) of fractured freeze-dried LLC-PKj cells. The cells in the top row were treated only with Cg-NBD-ceramide, the middle row represents cells treated with Cg-NBD-ceramide and then exposed to the Ca2+-free medium for 5 min in the presence of 2 mM EGTA, and the bottom row represents cells treated with Cg-NBD-ceramide and then exposed to 2 μM A23187 in Ca2+-free medium and 2 mM EGTA for 5 min. In ion microscope images, brightness indicates relative total calcium concentration within an image. The images of potassium, sodium and magnesium showed nearly homogeneous distribution throughout the cell (not shown). Calcium images were digitally processed and printed for optimum image quality. Ion image integration time for calcium images on the CCD imager was 100 s, Bar, 20 μm.

Quantitative, calibrated measurements of total calcium concentrations in three regions of the cell, the cell nucleus, the Golgi apparatus, and the rest of the cell cytoplasm (minus Golgi), for three cell lines are shown in Table 1. In all three cell lines, the Golgi apparatus contained significantly higher total calcium concentrations (per unit apparent volume) than the cell nucleus and the rest of the cell cytoplasm. It should be noted that a much higher, three-dimensional spatial resolution is required to resolve the calcium content of individual elements within the Golgi apparatus and to measure accurately organelle volume. Therefore, our measurements overestimate the Golgi volume and underestimate the actual local calcium concentration in the Golgi. However, the estimated distributions of total calcium amongst the three specified cellular regions are not significantly degraded by the resolution limitations. Based on the division of cell volume amongst these three regions as calculated by LSCFM, the Golgi apparatus contributes about 5% calcium in LLC-PK1 cells, 2.5% in 3T3 and L5 cells to the total cell calcium. The remainder of the cell cytoplasm is represented by an homogenized ion microscopic sampling of smaller cytoplasmic structures such as mitochondria and endoplasmic reticulum, and cytoskeleton and cytosol. The stated total calcium concentrations in this region are not directly comparable to calcium concentrations in distinguishable cytoplasmic structures. If all the cytoplasmic calcium were attributed to unresolved ER and mitochondria, which constitute fractions of cell volume comparable with Golgi, the intraorganelle calcium would also be comparable with the Golgi. Measurement uncertainties preclude specific figures here.

Table 1.

Total calcium concentrations and volume fractions of the nucleus, Golgi region and remaining cytoplasm

Total calcium concentrations and volume fractions of the nucleus, Golgi region and remaining cytoplasm
Total calcium concentrations and volume fractions of the nucleus, Golgi region and remaining cytoplasm

After 5 min of exposure of the LLC-PK1 cells to Ca2+-free medium (the middle row in Fig. 1), it is even easier to match the higher perinuclear calcium localization in each cell with the Golgi apparatus. The measurements indicate that calcium retention in the Golgi apparatus is significantly more robust than any other region of these cells. Treatment with A23187 (the bottom row in Fig. 1) released sequestered calcium from the Golgi more efficiently and reduced the calcium contrast amongst Golgi, and cytoplasm. Quantitative measurements of the calcium depletion experiments are summarized in Table 2. They confirm the visual estimates showing that the Golgi retains a higher percentage of calcium than any other region of the cell during calcium depletion. Measurements on the LLC-PKx cells showed continued loss of calcium during a longer exposure in Ca2+-free medium but again, at 15 min the Golgi apparatus retained a higher percentage of its initial total calcium (about 67%) as compared to the cell nucleus (about 43%), and the rest of the cell cytoplasm (about 33%).

Table 2.

% Total calcium retention in the nucleus, Golgi region and the rest of the cytoplasm in LLC-PKi, 3T3 and L5 cells upon exposure of cells to the calcium-free medium and A23187

% Total calcium retention in the nucleus, Golgi region and the rest of the cytoplasm in LLC-PKi, 3T3 and L5 cells upon exposure of cells to the calcium-free medium and A23187
% Total calcium retention in the nucleus, Golgi region and the rest of the cytoplasm in LLC-PKi, 3T3 and L5 cells upon exposure of cells to the calcium-free medium and A23187

In order to rule out the possibility that the C6-NBD-ceramide label may have affected the observed sequestration of calcium in the Golgi apparatus, the LLC-PKx cells grown on silicon chips in the same Petri dish were divided into two groups; one was stained with C6-NBD-ceramide and the other was not. In the control group (minus Ce-NBD-ceramide), similar perinuclear high calcium regions were observed in SIMS images. They were assumed to be localized in the Golgi apparatus and found to be directly comparable with the high calcium levels in the Golgi of the Cg-NBD-ceramide labeled group. No significant differences were observed in total calcium quantities between these labeled and unlabeled groups for any of the three examined regions of the cell. A 5 min exposure to the Ca2+-free medium of a control, unlabeled group resulted in levels of retention of total calcium in all three regions of the cell similar to those observed for the Cs-NBD-ceramide labeled group. These observations dem-onstrate that C6-NBD-ceramide does not detectably perturb intracellular calcium stores.

Although Cg-NBD-ceramide can also label mitochondria and endoplasmic reticulum, the fluorescence emanating from these structures was not distinguishable from background after 45 min post-staining incubation (Lipsky and Pagano, 1985a). The distribution of the mitochondria was determined by staining LLC-PKj cells with rhodamine 123 (Johnson et al. 1980). Fig. 2 shows a cryogenically prepared LLC-PKi cell stained to reveal mitochondria localization with LSCFM (top) and the corresponding total calcium SIMS image (bottom). A high density of mitochondria is distributed throughout the cell cytoplasm in this cell but not within the perinuclear calcium hot spot (arrow) associated with Golgi. Because the SIMS images could not distinguish the finely divided mitochondria and adjacent finely resolved ER, the distribution of total calcium amongst various stores in the cytoplasmic region could not be separated.

Fig. 2.

Rhodamine 123 LSCFM fluorescence image revealing mitochondria localization in a freeze-fractured, freeze-dried LLC-PKj cell (top) and the corresponding total calcium ion microscopy image (bottom). The arrows indicate a high-calcium perinuclear region. Bar, 10 μm.

Fig. 2.

Rhodamine 123 LSCFM fluorescence image revealing mitochondria localization in a freeze-fractured, freeze-dried LLC-PKj cell (top) and the corresponding total calcium ion microscopy image (bottom). The arrows indicate a high-calcium perinuclear region. Bar, 10 μm.

Identification of cytoplasmic structures with LSCFM in the same cell prior to SIMS analysis provides a new and powerful approach for recognizing structures with elemental content. The practical use of this method is enhanced by the simplicity of the cryogenic freeze-fracture method, which takes about 30 s to prepare a sample (Chandra et al. 1986). One can store frozen samples for a desired period of time under liquid nitrogen. The samples need to be freeze-dried prior to SIMS analysis. SIMS analysis is destructive in nature, since it is made by continuously eroding the cell surface; a 3-D mapping of the smaller structures in identical cells with a nondestructive technique like LSCFM is then ideal for microfeature recognition in SIMS images. An alternative to this approach would be a serial cryosectioning of the monolayer cultures, and the use of adjacent sections for transmission electron microscopy (TEM) and ion microscopy. The speed of such an analysis would be inferior to the present approach. The Golgi complex is composed of smooth membranes and its recognition in reliable unstained freeze-dried cryosections by TEM is problematic. This problem has limited the use of the electron probe to the elemental studies related to the Golgi complex. The three cell lines used in this study had a compact perinuclear Golgi complex, which could be resolved by the spatial resolution of the CAMECA IMS-3f ion microscope. A thinly spread Golgi apparatus may be difficult to resolve by this instrument. Similarly, the mitochondria and the endoplasmic reticulum could not be resolved.

Our observations have revealed a prominent perinuclear regional concentration of total calcium associated with the Golgi apparatus in three cell lines. The fractions of total cell calcium associated with the Golgi apparatus are only a few per cent in these cells. However, the robust retention of calcium by the Golgi during exposure to calcium-free medium suggests distinct calcium sequestration control. Several physiological roles for calcium sequestration in the Golgi can be anticipated. The Golgi apparatus may provide a pathway for removal of Ca2+ from cells by exocytosis of secretory vesicles originating in the Golgi. High concentrations of calcium in the Golgi may be required for aggregation of proteins in secretory vesicles (Gerdes et al. 1989). The Golgi apparatus is thought to play a role in intracellular Ca2+ homeostasis in human neutrophils (Krause and Lew, 1987). How the Golgi apparatus is involved in the agonist-induced release of Ca2+ remains to be determined.

This study also indicates that the cell nucleus has a large capacity to store calcium. The majority of this calcium has to be in the bound form otherwise intranuclear ionized Ca2+ will be very high. The binding of calcium to the nuclear material is possibly affected by calcium-free medium and A23187 treatments. The molecular species involved in intranuclear calcium binding are not clear but calcium-binding proteins may play a significant role (Schibeci and Martonosi, 1980).

Comparison of images of individual cells in which total calcium distributions are obtained by SIMS and organelle distributions are recorded by LSCFM of specific markers provides a powerful new approach for attribution of elemental content of various organelles and cellular structures under physiological and pathological conditions.

This work was supported by NIH GM 24314 and DOE DE-FG02-91ER61138 grants to G.H.M., and NIH 5P41-RR-04224 and NSF DIR-8800228 grants to W. W. W. The Cornell NIH/NSF Developmental Resource for Biophysical Imaging and Optoelectronics was used in culturing the cells and fluorescence imaging. Assistance of K. S. Wells and D. Sandison is acknowledged in fluorescence imaging experiments. D. Sandison is also thanked for a constructive criticism of the manuscript.

Ausserer
,
W. A.
,
Chandra
,
S.
and
Morrison
,
G. H.
(
1988
).
Morphological and elemental integrity of freeze-fractured, freeze-dried cultured cells during ion microscopic analysis
.
J. Microsc. (Oxford
.
154
,
39
57
Ausserer
,
W. A.
,
Ling
,
Y-C.
,
Chandra
,
S.
and
Morrison
,
G. H.
(
1989
).
Quantitative imaging of boron, calcium, magnesium, potassium and sodium distributions in cultured cells with ion microscopy
.
Analyt. Chem
.
61
,
2690
2695
.
Berridge
,
M. J.
and
Irvine
,
R. F.
(
1989
).
Inositol phosphates and cell signalling
.
Nature
341
,
197
205
.
Cameron
,
I. L.
,
Sparks
,
R. L.
and
Sellig
,
L. L.
, Jr
. (
1980
).
Concentration of calcium and other elements at a subcellular level in the lactating epithelium of rat
.
Cytobios
27
,
89
96
.
Castaing
,
R.
and
Slodzian
,
G.
(
1962
).
Microanalyse par emission ionique secondaire
.
J. Microscopie (Paris
.
1
,
395
410
.
Chandra
,
S
,
Ausserer
,
W. A.
and
Morrison
,
G. H.
(
1987
).
Evaluation of matrix effects in ion microscopic analysis of freeze-fractured, freeze-dried cultured cells
.
J. Microsoc. (Oxford)
148
,
223
239
.
Chandra
,
S.
and
Morrison
,
G. H.
(
1985
)
Imaging elemental distribution and ion transport in cultured cells with ion microscopy
.
Science
228
,
1543
1544
.
Chandra
,
S.
and
Morrison
,
G. H.
(
1988
).
Ion microscopy in biology and medicine
.
Methods Enzymol
.
158
,
157
179
.
Chandra
,
S.
,
Morrison
,
G. H.
and
Wolcott
,
C. C.
(
1986
).
Imaging intracellular elemental distribution and ion fluxes in cultured cells using ion microscopy: A freeze-fracture methodology
.
J. Microsc. (Oxford
.
144
,
15
37
.
Freedman
,
R. A.
,
Maclaughlin
,
J. A.
and
Weiser
,
M. M.
(
1981
).
Properties of Ca2+ uptake and release by Golgi membrane vesicles
.
Archs Biochem. Biophys
.
206
,
233
241
.
Gerdes
,
H-H.
,
Rosa
,
P.
,
Phillips
,
E.
,
Baeuerle
,
P. A
,
Frank
,
R.
,
Argos
.
P.
and
Huttner
,
W. B.
(
1989
).
The primary structure of human secretogranin II, a widespread tyrosine-sulfated secretory granule protein that exhibits low pH- and calcium-induced aggregation
J. biol. Chem
.
264
,
12 009
12015
.
Ghosh
,
T. K.
,
Mullaney
,
J. M.
,
Tarazi
,
F. I.
and
Gill
,
D. L.
(
1989
).
GTP-activated communication between distinct inositol 1,4,5-trisphosphate-sensitive and -insensitive calcium pools
.
Nature
340
,
236
239
.
Hodson
,
S.
(
1978
).
The ATP-dependent concentration of calcium by a Golgi apparatus-rich fraction isolated from rat liver
.
J. Cell Sci
.
30
,
117
128
.
Johnson
,
L. V.
,
Walsh
,
M. L.
and
Chen
,
L. B
(
1980
).
Localization of mitochondria in living cells with rhodamine 123
.
Proc natn Acad. Sci. U.S.A
.
77
,
990
994
.
Krause
,
K.-H.
and
Lew
,
P. D.
(
1987
).
Subcellular distribution of Ca2+ pumping sites in human neutrophils
.
J. clin. Invest
.
80
,
107
116
.
Ling
,
Y.-C.
,
Bernius
,
M. T.
and
Morrison
,
G. H.
(
1987
)
SIMIPS Secondary ion mass image processing system
.
J. Chem. Inf. Comput. Sci
.
27
,
86
94
.
Lipsky
,
N. G.
and
Pagano
,
R E.
(
1985a
).
A vital stain for the Golgi apparatus
.
Science
228
,
745
747
.
Lipsky
,
N. G.
and
Pagano
,
R E.
(
1985b
).
Intracellular translocation of fluorescent sphingolipids in cultured fibroblasts: Sphingomyelin and glucocerebroside analogues pass through the Golgi apparatus en route to the plasma membrane
.
J. Cell Biol
.
100
,
27
34
.
Mantus
,
D. S.
and
Morrison
,
G. H.
(
1990
)
Ion image detection with a microchannel plate evaluated by using a charge coupled device camera
Analyt. Chem
.
62
,
1148
1155
.
Meldolesi
,
J.
,
Madeddu
,
L.
and
Pozzan
,
T.
(
1990
).
Intracellular Ca2+ storage organelles in non-muscle cells: heterogeneity and functional assignment
.
Biochim. biophys. Acta
1055
,
130
140
Morrison
,
G. H
and
Slodzian
,
G.
(
1975
).
Ion microscopy
.
Analyt. Chem
47
,
932A
943A
.
Neville
,
M. C.
,
Selker
,
F.
,
Semple
,
K.
and
Watters
,
C.
(
1981
).
ATP-dependent calcium transport by a Golgi-enriched membrane fraction from mouse mammary gland
.
J. Membr. Biol
,
61
,
97
105
.
Pagano
,
R. E.
and
Sleight
,
R G.
(
1985
).
Defining lipid transport pathways in animal cells
.
Science
229
,
1051
1057
.
Roos
,
L
(
1988
).
A possible site of calcium regulation in rat exocrine pancreas cells: An X-ray microanalytical study
.
Scanning Microsc
.
2
,
323
329
.
Schibeci
,
A.
and
Martonosi
,
A.
(
1980
).
Ca2+-Binding proteins in the nuclei
.
Eur. J. Biochem
.
113
,
5
14
.
Snedecor
,
G. W.
and
Cochran
,
W. G.
(
1989
).
Statistical Methods (8th edn
.,
Iowa State University Press, Ames, Iowa
, pp.
324
329
.
Somlyo
,
A. P.
,
Bond
,
M
and
Somlyo
,
A. V.
(
1985
).
Calcium content of mitochondria and endoplasmic reticulum in liver frozen rapidly in vivo
Nature
314
,
622
625
.
Wells
,
K. S.
,
Sandison
,
D R.
,
Strickler
,
J
and
Webb
,
W. W.
(
1990
).
In Quantitative Fluorescence Imaging with
.
Laser Scanning Confocal Microscopy
,
Handbook of Biological Microscopy, Plenum 1990
,
New York
.
West
,
D. W.
(
1981
).
Energy-dependent calcium sequestration activity in a Golgi apparatus fraction derived from lactating rat mammary glands
.
Biochim. biophys. Acta
673
,
374
386
.
White
,
J. G.
,
Amos
,
W. B.
and
Fordham
,
M.
(
1987
).
An evaluation of confocal versus conventional imaging of biological structures by fluorescence light microscopy
.
J. Cell Biol
.
105
,
41
48
.