ABSTRACT
Calcium transport from the eggshell to the developing chick embryo is carried out by the ectoderm cells of the chick chorioallantoic membrane. Primary cells isolated from chick chorioallantoic membrane ectoderm were used to analyze the subcellular distribution of 45Ca2+ accumulated from the extracellular medium. We present evidence suggesting that calcium may be sequestered into endosome-like vesicles during the initial phase of uptake. A combination of techniques were utilized to monitor calcium fluxes and calcium compartmentalization in the cultured chorioallantoic membrane cells: (1) fura-2 fluorescence was used to indicate cytosolic free calcium concentrations, (2) 45Ca2+ tracer was used to follow calcium accumulation in all cellular compartments, and (3) digitonin was used to differentially permeabilize subcellular membranes in order to localize 45Ca2+ by following tracer release profiles.
Differences between cytosolic calcium flux and whole cell calcium accumulation suggested that the pathway of calcium uptake from the medium involves sequestration into an internal compartment separate from the cytosol. Kinetic analysis of the digitonin-mediated release of specific subcellular markers (lactate dehydrogenase, NAD-dependent isocitrate dehydrogenase, [3H]inulin, and [3H]-2-deoxyglucose) and preloaded 45Ca2+ indicated that calcium was localized in a compartment similar to endosomal vesicles. Our results are consistent with a transcytotic mechanism for chorioallantoic membrane calcium transport.
INTRODUCTION
Calcium is an important nutritional requirement for vertebrates; it is required for a large number of diverse activities including the formation and maintenance of calcified tissues, the modulation of cellular activities (e.g. muscle contraction), certain types of cell adhesion, complex signal transduction pathways, apoptosis, etc. In vertebrates, systemic calcium levels are modulated by specific tissue types, including: intestine, which supports bidirectional calcium transport (Sheikh et al., 1990); kidney, which modulates calcium excretion (van Os, 1987); bone, where hydroxyapatite deposition, remodeling and resorption take place (Mundy, 1989). During the development of most mammals, the placenta serves as the tissue providing calcium during development; on the other hand, in oviparous species the calcium supply function of the placenta is accomplished by the embryonic chorioallantoic membrane (CAM).
Some tissues are capable of rapid transport of large quantities of calcium; for example, the chicken shell gland supplies nearly 50 mmoles of calcium to the forming eggshell in a period of approximately 16 hours (Burton, 1987); a square centimeter of chick CAM transports in excess of 100 nmoles of calcium per hour (Crooks and Simkiss, 1975). This prodigious transport is carried out by epithelial cells which, while involved in the translocation of large amounts of calcium, maintain very low cytosolic Ca2+ concentrations ([Ca2+]i). Although the specific mechanism of calcium handling during transport is, in many cases, unclear, four general schemes have been suggested: (1) calcium transport may involve influx across the apical plasma membrane, cytosolic transit, which may involve a calcium binding protein to keep [Ca2+]i low and (acting through mass action) to help translocate calcium and, finally, removal of Ca2+ through a basolateral pumping mechanism (see Bronner, 1990); (2) calcium transport may occur via an extracellular passive diffusion mechanism where Ca2+ accumulates at the cell surface and transport occurs along cellular interstices (Saleuddin et al., 1976; van Os, 1987); (3) calcium transporting cells may utilize subcellular organelles, particularly mitochondria, to sequester large amounts of cellular calciumto be released into the cytosol in a controlled fashion, and then pumped across the basolateral membranes (Borle, 1973); (4) calcium may be transported by a modified endocytosis-like process involving calcium accumulation at the apical surface of the cells, followed by internalization and vesicle-mediated delivery to the basolateral surface (Terepka et al., 1976; Tuan, 1987). Recently, bio-chemical analyses have implicated vesicular calcium transport by the intestinal mucosa (Nemere and Norman, 1990); however, direct experimental evidence concerning which mechanism(s) may be used by particular transport epithelia to handle calcium transport is still lacking.
In the preceding paper, we showed that primary cells isolated from the CAM of the chicken (Gallus gallus ) exhibited 45Ca2+ uptake kinetics similar to those seen for CAM tissue discs in vitro (Stewart and Terepka, 1969; Garrison and Terepka, 1972a,b; Terepka et al., 1976) and reported for the intact CAM in ovo (Crooks and Simkiss, 1975). Curve fitting of uptake data indicated that calcium was being accumulated into a single cellular compartment. In order to assess the cellular fate and distribution of the calcium accumulated by CAM cells, we prepared cultures for in vitro analysis and traced calcium to particular cellular compartments. Since CAM ectoderm cells have relatively simple morphology (Coleman and Terepka, 1972c; also see preceding paper), the compartments available for calcium accumulation were limited; i.e. the cell surface, cytosol, mitochondria, and subcellular vesicles. In this study, using several parallel lines of investigation, cultured CAM ectoderm cells were used to identify the subcellular distribution of Ca 2+ taken up from the extra-cellular medium.
In one set of experiments, 45Ca2+ was used as a tracer for calcium accumulation. The time course of 45Ca2+ uptake represented the exchange of intracellular calcium stores with extracellular calcium. For several cell types (Borle, 1990), calcium exchange occurs over the course of many hours; however, for CAM preparations, in ovo and in vitro, calcium exchange occurs through a characteristic, though as yet unidentified, rapid influx pathway. By comparing the cellular accumulation of calcium with the accumulation of calcium in the cytosol, as measured by fura-2 ratio fluorescence, we sought to assess the degree to which the incoming calcium was sequestered. Kinetic and steady state comparisons were done to determine whether the total calcium influx could be accounted for by cytosolic calcium accumulation alone or if a non-cytosolic compartment was involved in the rapid accumulation of calcium from the medium.
Additional experiments were performed to determine the intracellular localization of calcium accumulated by isolated CAM ectoderm cells. Cells were incubated with combinations of compartment-specific, 3H-labeled markers and/or 45Ca2+. Radiolabeled materials were released into the medium by the selective and progressive permeabilization of cellular membranes with various concentrations of digitonin (DGT) or Triton X-100 (Tx-100). Since DGT has been shown to differentially permeabilize cell membranes (Diaz and Stahl, 1989), comparisons of the release profiles for 45Ca2+ with the release profiles of other radiolabeled markers, as well as characteristic enzymatic activities, from known cellular compartments allowed the assignment of a particular subcellular localization for 45Ca2+ taken up from the external medium. Our results lead us to propose that the calcium accumulation phase of calcium transport involves compartmentalization of calcium into endosome-like, vesicular structures. Our observations are consistent with the vesicle-mediated model of transcellular calcium transport.
MATERIALS AND METHODS
Isolation and culture of CAM ectoderm cells
Cells were isolated and cultured as described in the preceding paper. Briefly, the ectodermal cell layer of the CAM was separated from the adherent inner shell membrane by hydration, and endodermal cells were removed by stripping them away onto nitrocellulose. The remaining tissue was then selectively digested with trypsin and EDTA, for the preferential isolation of ectodermal cells. After removal of contaminating fibroblasts by brief pre-plating, the CAM ectodermal cells were plated on Matrigel (1:10 dilution; Collaborative Research) and cultured at 37°C with a 5% CO2 atmosphere, in Ham’s F-12 medium containing 10% calf serum and 5 μM hemin.
Calcium uptake by isolated CAM ectoderm cells
Incorporation of calcium by isolated CAM ectoderm cells was carried out as described in the preceding paper. Briefly, calcium was added to a cell suspension (final concentration of 0.1 or 1.0 mM as indicated) and, at time zero, 45Ca2+ was added as a tracer. At various time points, samples of the cell suspension were collected by centrifugation through silicone oil, and 45Ca2+ accumulation was determined by liquid scintillation counting. In some experiments, the harvested cells were exposed to either 1 mM EGTA or 2 mM CaCl2 to remove, or compete away, external 45Ca2+ label prior to collection. The amount of calcium accumulated was determined, based on the specific activity of 45Ca2+ in the bathing medium.
Calcium concentration measured by fura-2 ratio fluorescence
fura-2 measurements (Grynkiewicz et al., 1985) were carried out on a Leitz Fluovert microscope retrofitted with UV transmissive optics in the epi-illumination pathway and an Olympus ×40 PlanApoUV objective. A PTI Delta-Scan system (Photon Technology International) was used for the production and analysis of fura-2 fluorescence signals. Cells were cultured on acid-ethanol washed, Matri-Gel coated, no.0 22 mm2 glass coverslips and allowed to recover overnight. They were subsequently depleted of cellular calcium stores by exposure to Hanks’ Balanced Salt Solution containing 10 mM Hepes-NaOH, pH 7.2, (HBSSH) with added 1-5 mM EGTA. Calcium depletion took approximately 30 min, as monitored by fura-2 fluorescence, although the EGTA exposure time varied depending on the cell selected. Cultures were ‘bead-loaded’ with 10 μM fura-2 pentapotassium salt in buffered saline as previously described (McNeil et al., 1985; Miyahara et al., 1992). Perfusion chambers were constructed, using spacers made of chips of no.0 coverslips, by inverting the coverslips onto acid-ethanol-washed microscope slides so that opposite corners would overhang. The chambers were sealed with a 1:1 mixture of melted paraffin wax and Vaseline which was applied along the edges of the coverslip. When the chambers were inverted onto the microscope stage, the protruding coverslip corners provided convenient ports for the perfusion of different bathing solutions, which could easily be removed using a filter paper wick. Cells were selected for analysis based on three criteria: (1) even fluorescence and recognizable morphology; (2) adherence to the coverslip during perfusion; and (3) stable single channel fura-2 fluorescence signal indicating that the dye was not leaking from the cell. A field diaphragm allowed individual cells to be selectively monitored. Excitation wavelengths were determined empirically by comparing wavelength scans in Br-A23187-permeabilized cells in the presence of 1 mM CaCl2 or 1 mM EGTA. Values of Rmax, Rmin and the single wavelength fluorescence ratio for free versus bound dye (SFB) were determined in the Br-A23187-permeabilized cells. Time course measurements were done at excitation wavelengths of 352 and 380 nm for the determination of [Ca2+]i, and fura-2 fluorescence levels were monitored continuously through a 495 nm long pass emission filter. Background levels from system scatter, reflectance and black level were subtracted from raw fluorescence data prior to ratio calculation; in general, cellular auto-fluorescence was negligible.
The Leitz Fluovert and PTI Deltascan were also used for the determination of Ca2+ concentrations in vitro. fura-2 pentapotassium salt (Molecular Probes, Eugene, Oregon) was dissolved in 150 mM KCl with 10 mM HEPES (pH 7.2) to a concentration of 10 mM. Transcalcin extract was prepared as described by Tuan et al. (1978) and dialyzed against 150 mM KCl with 10 mM HEPES (pH 7.2). The resulting solution was dispersed by adding 1% Tx-100. The transcalcin preparation was combined with fura-2 pentapotassium salt so that the dye concentration was 100 μM, and the changes in fluorescence levels caused by the addition of Ca2+ were determined photometrically. Values of Rmax, Rmin and SFB were determined separately for each dilution of extract.
Transcalcin, the major Ca-binding protein of the CAM, binds Ca2+ with high affinity. The transcalcin/Ca2+ complex has a Kd of 2.35×10−7 M (Tuan and Scott, 1977), which is approximately equivalent to the Kd for a fura-2/Ca2+ complex (2.24×10−7 M) (Grynkiewicz et al., 1985). It was possible, therefore, to use fura-2 fluorescence to indirectly follow the interaction between Ca2+ and transcalcin both in vitro and in vivo. Specifically, by monitoring the fura-2 ratiometric fluorescence change in the presence of transcalcin in vitro, the kinetic aspects of the transcalcin/Ca2+ interaction were determined. These characteristics were then used to interpret cytosolic calcium fluxes and cellular calcium uptake to assess the possibility of a transcalcin sink in the CAM transport mechanism.
Differential permeabilization of CAM cells
Cellular membranes were differentially permeabilized with various concentrations of digitonin (DGT) after the method of Diaz and Stahl (1989). These workers showed that (1) at low concentrations, DGT selectively permeabilizes plasma membranes, and (2) at higher concentrations DGT also breaks down internal membranes (endosomes, lysosomes, endoplasmic reticulum, Golgi apparatus, etc.) but leaves inner mitochondrial membranes intact. Cells were isolated and cultured as described above. For the determination of released enzyme activity, cells were rinsed with HBSS then permeabilized with digitonin (DGT) dissolved in ‘Cytoplasmic Buffer A’ (10 mM HEPES, 0.8 mM MgCl2, 2.4 mM KCl, 12 mM NaHCO3, 130 mM CH3CO2K, brought to pH 7.2 with KOH) at 37°C for the indicated time periods. The enzyme activity per unit volume of the non-cellular supernatant was determined and compared to the whole culture activity released by solubilizing the cells in Triton X-100.
For the determination of the release of radioactive markers, cells were incubated at 37°C for 5 min in the presence of radioactive 45Ca2+, [3H]inulin (a fluid-phase marker), and/or [3H]-2-deoxyglucose (a cytosolic marker). The markers were chased for an additional 0.5 to 5 min with unlabeled medium. The cells were then permeabilized with various concentrations of DGT or Tx-100. Time courses of DGT-mediated release were compared to the total release by subsequent Tx-100 solubilization. All radioactivity levels were obtained as d.p.m. by liquid scintillation counting.
Enzyme assays
Lactate dehydrogenase (LDH) activity levels were determined spectrophotometrically utilizing a kit from Sigma (LDL-10). Samples were collected as cell supernatants released by DGT or Tx-100 treatment in ‘Cytoplasmic Buffer A’. Mitochondrial NAD-dependent isocitrate dehydrogenase (ICD) activity was similarly determined spectrophotometrically using a modification of a kit from Sigma (176-C). NAD was substituted for the supplied NADP substrate, and 1 mM Na-ADP was added to ‘Cytoplasmic Buffer A’ for use in DGT- and Tx-100-mediated solubilization of ICD samples.
RESULTS
Internalization of calcium by isolated CAM cells
We first determined the extent to which calcium uptake by isolated CAM cells indicated internalization of calcium as opposed to cell surface accumulation. Resuspended cells were incubated with 0.1 mM CaCl2 containing tracer 45Ca2+ for 0.5 and 2 min, followed by the addition of excess unlabeled CaCl2 (2 mM) or EGTA (1 mM) before collection. Cellular 45Ca2+ incorporation was determined, and the levels compared to control cells that were not exposed to EGTA or excess unlabeled CaCl2. As shown in Fig. 1, neither EGTA nor excess CaCl 2 had a substantial effect on the initial rate of 45Ca2+ uptake determined by the difference in label accumulated between 0.5 and 2 min. The results indicated that most of the 45Ca2+ accumulated during the uptake phase of transport was not accessible to external binding by EGTA or exchange with CaCl2 and was, therefore, internalized by the isolated CAM ectoderm cells.
Cytosolic calcium flux and cell calcium uptake
We next analyzed the mechanism of calcium internalization by isolated CAM cells. One possibility is that calcium accumulates in the cytosol where a calcium binding protein (CaBP) may act to keep the effective [Ca2+]i low and to facilitate mass transcellular movement of the ion. To assess this possibility, we applied fura-2 ratiometric fluorimetry to single cells. We first examined the degree to which transcalcin might alter the kinetic profile of fura-2 fluorescence by utilizing an in vitro, cell-free preparation containing transcalcin derived from CAM extracts as described in Materials and Methods. Fig. 2 shows that the absolute baseline and plateau levels of fura-2 fluorescence correlated with the amount of transcalcin extract present, but a rapid increase in signal was seen with the addition of Ca2+ in each case. The slope of this signal increase was apparently independent of the amount of transcalcin present, indicating that transcalcin bound to Ca2+ relatively rapidly. As a result, changes in [Ca2+]i would be directly reflected by cytosolic fura-2 fluorescence and only the magnitude of those changes would be affected by the presence of transcalcin.
To measure calcium fluxes in CAM ectoderm cultures, cells were first depleted of cellular calcium by prolonged exposure to EGTA (to unload the presumptive cytosolic Ca2+ pool). A decrease in [Ca2+]i, as monitored by fura-2 fluorescence, from 200-300 nM down to 50-100 nM, indicated that this treatment was effective. When Ca2+ was added back to the medium, [Ca2+]i was measured by fura-2 fluorescence while, in parallel experiments, total Ca2+ accumulation was monitored by 45Ca2+ tracer analysis. Our results clearly show that the profiles of whole cell calcium accumulation (45Ca2+ measurements) and cytosolic Ca2+ flux (fura-2 measurements) differ dramatically. Fig. 3 shows typical fura-2 kinetic profiles for single CAM ectoderm cells at 37°C and room temperature (approx. 23°C). [Ca2+]i rose rapidly from a baseline of 90 nM to a peak value (approx. 1.8 μM) within 1 to 2 min then decreased to a plateau level (300 nM) within 10 to 15 min. There were no major differences between the profiles obtained at the two temperatures. Fig. 4, on the other hand, shows the time course of overall cellular calcium accumulation, measured by 45Ca2+ uptake. As reported in the accompanying paper for CAM tissue in ovo, tissue discs in vitro, and non-EGTA treated ectoderm cells in vitro, cellular calcium levels rose smoothly to a steady state level over a 5 to 10 min period in a temperature-dependent manner. The profiles of changes in [Ca2+]i and overall calcium accumulation were, therefore, kinetically different and apparently represented different calcium pools.
Compartmentalization of calcium in CAM ectoderm cells
We also examined the extent to which 45Ca2+ accumulated in the available subcellular compartments of isolated CAM ectoderm cells. We tested the applicability of DGT-mediated selective permeabilization of CAM ectodermal cell membranes for use in detecting calcium stored in membranous organelles. Fig. 5 shows the relative release of two cellular enzymes after exposure to DGT for 20 min at 4°C: (A) LDH, a marker for the cytosol, to indicate permeabilization of the plasma membrane; and (B) NAD-dependent ICD, a marker for mitochondria, to indicate permeabilization of the inner mitochondrial membrane. The release profiles of LDH and ICD show that increasing DGT concentration increased LDH release, but that even high levels of DGT did not release ICD. By comparison, Triton X-100 solubilization released high levels of both enzymes. These findings indicate that DGT was effective in the preferential permeabilization of CAM cell plasma membranes over mitochondrial membranes.
We next determined the degree to which DGT preferentially solubilized the plasma membrane of CAM ectoderm over internal vesicle/endosome membranes. Plasma membrane permeability was assessed first: cells were preloaded with [3H]-2-deoxyglucose, which accumulates in the cytosol, external [3H]-2-deoxyglucose was rinsed away, and internalized [3H]-2-deoxyglucose was released by treatment with various concentrations of DGT. Fig. 6 shows the time course of [3H]-2-deoxyglucose release at 4°C, and demonstrates the concentration dependence of plasma membrane integrity in the presence of DGT. In particular, 0.5 μM DGT did not release [3H]-2-deoxyglucose even after 30 min of incubation, whereas substantial release was seen shortly after incubation with 10 μM DGT. These results further supported the usefulness of DGT for the selective permeabilization of CAM cells.
Since the results of our kinetic analyses had indicated that Ca2+ was sequestered away from the cytosol, we determined whether the time course for the release of a fluid-phase marker ([3H]inulin) from endosomes could be distinguished from the release of a cytosolic marker ([3H]-2-deoxyglucose). Fig. 7A shows the release of radiolabeled inulin and 2-deoxyglucose from CAM ectoderm cells by solubilization with 10 μM DGT at 37°C. The time courses were clearly distinguishable, with inulin release lagging behind 2-deoxyglucose release, as would be expected from its subcellular compartmentalization in endosomes rather than the cytosol.
Finally, we investigated the localization of calcium shortly after uptake by examining the kinetics of 45Ca2+ release from CAM ectoderm cells as a result of the differential permeabilization of subcellular membranes by DGT. Fig. 7B shows a time course for 45Ca2+ release in the presence of 10 μM DGT at 37°C. By comparison, the release of 45Ca2+ was similar to the release of [3H]inulin but lagged behind the release of [3H]-2-deoxyglucose from the cytosol. The relationship between 45Ca2+ release and [3H]inulin release was similar for samples chased for as little as 30 s to as long as 5 min, as shown in Fig. 7C, indicating that the release kinetics were not due to 45Ca2+ label leaking from a sequestered store into the cytoplasm prior to permeabilization. Taken together, the results of this study (Figs 5-7) are consistent with calcium being stably sequestered in an endosome-like compartment in the CAM ectoderm cells shortly after uptake.
DISCUSSION
Calcium transport by the chick CAM is a regulated process that supplies the developing embryo with eggshell calcium to be used in the formation of bones. Transepithelial transport by the CAM has been shown to be active and unidirectional (Garrison and Terepka, 1972), but the mechanism has been only partially characterized. The ability to prepare cultures of CAM ectoderm cells (see the accompanying paper) has allowed us to analyze directly the early events of CAM calcium transport and to distinguish among several possible transport mechanisms.
The accumulation of calcium by isolated CAM ectoderm cells is not typical of cells in general but occurs rapidly and specifically at near physiological temperatures. It should be pointed out that the calcium influx rate for CAM ectoderm cells (which ranged from 0.5 to 6.5 nmole/106 cells per min, depending on the preparation, see Fig. 4) was significantly higher than the corresponding rates for cytosolic calcium accumulation in cells isolated from other calcium transporting epithelia (25- and 40-fold higher than primary chick duodenal cells and LLC-MK2 kidney cells, respectively; see Borle, 1974). In addition, in vitro CAM ectoderm cells have calcium influx rates 4 to 5 orders of magnitude faster than the net permeability of the human erythrocyte in the absence of Ca-pump extrusion activity (see Lew, 1990). The comparison of calcium influx rates among these cell types indicates a specialized calcium entry pathway in the CAM. In addition, as seen in the plateau region of Fig. 3, resting CAM ectoderm cells contain approximately 200 nM Ca2+, a value that is higher than the amount of resting Ca2+ in other cell types and may be related to the specialized function of these cells. Since isolated CAM cells maintain the high rate of calcium uptake typical of the apical surface of the CAM ectoderm, determination of the mechanisms involved in CAM cell calcium homeostasis should be relevant to the mechanism of CAM calcium transport.
We have shown that 45Ca2+ accumulation by isolated CAM ectoderm cells involves actual internalization since the uptake rate was not altered by the addition of EGTA or excess nonradioactive Ca2+, as either treatment would be expected to lower the amount of externally associated 45Ca2+ without affecting the level of internalized marker. In light of the suggestions of Wasserman et al. (1974) as extended by many others (see Bronner, 1990), it seems possible that calcium accumulated in the cytosol of the CAM ectoderm cells bound to a high affinity CaBP. The chick CAM possesses a high affinity CaBP (transcalcin) that accounts for nearly all of the calcium-binding activity of extracts (Tuan and Scott, 1977), and we therefore examined the nature of transport calcium internalization by parallel measurements of cytosolic fura-2 fluorescence and 45Ca2+ uptake; however, since the 45Ca2+ and fura-2 kinetic profiles have different shapes and different responses to decreased temperature, it appeared unlikely that calcium accumulation resulted directly and exclusively from increased cytosolic traffic. It is important to note that this conclusion could not have been drawn without the in vitro characterization of transcalcin-containing extracts; specifically, if transcalcin bound Ca2+ slowly, relative to fura-2, then the presence of transcalcin would have become significant when interpreting the shape of the cellular fura-2 kinetic profiles.
In addition to differences in the kinetic profiles between cellular and cytosolic calcium uptake, there were differences in the level of whole cell calcium accumulation and the level of cytosolic calcium accumulation. The change in steady state [Ca2+]i indicated by fura-2 measurements allowed a calculation to be done to determine the amount of transcalcin necessary to account for the cellular uptake levels indicated by 45Ca2+ measurements. Based on the gross surface area of the CAM (58 cm2; Romanoff, 1961), the average area of CC cells (45-50 μm2, see preceding paper), the peak amount of transcalcin present in a mature CAM (12.4 μg, or 1.24 nmole of binding sites per CAM; Tuan and Scott, 1977), the average volume of freshly isolated CAM ectoderm cells (approximately 0.52 pl, based on Coulter Counter data; not shown), and assuming a diffuse distribution for transcalcin, a generous estimate of 20 μM can be made for the transcalcin concentration in CAM ectoderm cells. Since the Kd for Ca2+/transcalcin is 2.35×10−7 M (Tuan and Scott, 1977), the proportion of transcalcin bound at the baseline and plateau cytosolic Ca2+ concentrations given in Fig. 3 can be estimated: at [Ca2+]free = 90 nM, 28% would be bound, and at [Ca2+]free = 300 nM, 56% would be bound. A Ca2+ uptake of approx. 0.5 nmole/106 cells indicates an exchangeable, steady-state pool of 962 μM calcium; this concentration accumulated in the cytosol would require 1.7 mM transcalcin, an amount 85-fold higher than our already generous estimate. In fact, the disparity between the transcalcin available and that required for a cytosolic accumulation model would be much greater since at day 13 only about 20% of the peak transcalcin level would be present (Tuan, 1980). It is clear that, although CAM ectoderm cells quickly internalize large amounts of calcium, this calcium, by quantitative necessity, must be sequestered away from the cytosol.
As shown in previous studies, the cells located between the calcium-rich inner shell membrane and the embryonic vasculature become terminally differentiated in ovo and lose biosynthetic machinery (Coleman and Terepka, 1972a,b). In their differentiated state, the cells have few organelles available for calcium sequestration. If calcium were stored in mitochondria, as suggested by Borle (1973) for kidney cells, the results of the DGT release experiments would be difficult to explain for two general reasons: (1) calcium that was stably sequestered in mitochondria would not have been released by DGT permeabilization; and (2) in the event that calcium sequestered by mitochondria was able to exchange with the cytosolic calcium pool and was, therefore, releasable by DGT, the release would have been dependent on the duration of the chase phase of the experiment. As shown by Figs 5-7, the observed release kinetics of 45Ca2+ closely resembles the release of [3H]inulin and, therefore, the DGT release experiments indicate that 45Ca2+, a marker for actively accumulated calcium was, most likely, stably sequestered in an endosome-like compartment of isolated CAM ectodermal cells.
The transcytosis model initially proposed by Terepka et al. (1976) and modified by Tuan (1987), (also see Tuan et al., 1991) involves the endocytic uptake of solubilized eggshell calcium at the apical surface of the CAM ectodermal CC cells. The results presented here are consistent with a transcytosis model. In addition, Dunn and Fitzharris (1987) have demonstrated active endocytosis by the CAM ectoderm in situ, and Nemere and Norman (1990) have implicated endosomal/lysosomal compartmentalization of transport calcium in the chick intestine. We are currently in the process of further delineating the mechanistic steps involved in the development-specific transepithelial translocation of calcium by the CAM.
ACKNOWLEDGEMENTS
This work was supported in part by grants from the NIH (HD15822 and HD21355), March of Dimes Birth Defects Foundation (1-1146), USDA (88-37200 and 90-37200-5265), and the Orthopaedic Research and Education Foundation.