Localization of calcium in vas deferens using 45Ca EM autoradiographyRelationship to species and the effect of 45Ca removal.код для вставкиСкачать
THE ANATOMICAL RECORD 217:321-327 (1987) Localization of Calcium in Vas Deferens Using 45Ca EM Autoradiography: Relationship to Species and the Effect of 45CaRemoval LINDA J. McGUFFEE, SALLY A. LITTLE, AND BETTY J. SKIPPER Department of Pharmacology (L.J. M., S . A . L ) and Department of Family, Community, and Emergency Medicine (B.J.S.), School of Medicine, University of New Mexico, Albuquerque, NM 87131 ABSTRACT The distribution of calcium was determined in the vas deferens of the guinea pig using 45Ca electron microscopic autoradiography of rapidly frozen, freeze-dried, and embedded tissue. A selective accumulation of calcium at the plasma membrane and SR was observed in vas deferens that had been incubated in 45Ca for 65-85 min prior to rapid freezing. Rinsing the tissue in nonradioactive calcium for 6 min prior to rapid freezing significantly altered the distribution of calcium among the plasma membrane, mitochondria, and cytoplasmic matrix. The influence of species on the observed distribution of calcium was also examined. The distribution of calcium in the guinea pig vas deferens was not significantly different from that in the rabbit vas deferens when the tissues were prepared under identical conditions. The role of calcium in smooth muscle contraction has been studied extensively, and both extracellular and intracellular pools of activator calcium have been identified (Daniel et al., 1983; Hudgins and Weiss, 1968; Johansson and Somlyo, 1980; Van Breemen et al., 1980). However, the ultrastructural locations of the extracellular or intracellular activator calcium have not been determined. It has been postulated that the clear, membrane-bound spaces (presumably sarcoplasmic reticulum), mitochondria, and plasma membrane may serve as sites of intracellular calcium sequestration (Daniel et al., 1983; Johansson and Somlyo, 1980; McGuffee et al., 1981; Popescu and Diculescu, 1975; Sugi and Daimon, 1977; Suzuki, 1982), although the relative contribution of these sites as sources of activator calcium has only recently begun to be investigated using morphological techniques (Bond et al., 1984a; McGuffee et al., 1985). Our laboratory has been examining the usefulness of 45Ca autoradiography (ARG) at the electron microscopic (EM) level of resolution in evaluating calcium distributions in smooth muscle. Two requirements must be met before one can examine the distribution of a soluble ion such as calcium. First, tissue ultrastructure must be sufficiently preserved so that intracellular organelles can be identified. Second, the distribution of calcium observed must be representative of the in vivo distribution of the ion. We have previously shown that the first requirement can be satisfied by quick freezing followed by freeze drying (McGuffee et al., 1979; McGuffee et al., 1981; McGuffee et al., 1985).We have also examined the extraction of calcium during tissue processing (McGuffee et al., 1981; McGuffee and Little, 1986). Those studies indicate that extraction of calcium is greatly reduced in freeze-dried tissue compared to conventionprepared tissue. The that is can be associated with the removal Of extracellular 01 SIXface calcium. We must emphasize that redistribution 0 1987 ALAN R. LISS, INC. within the tissue has not been measured directly. However, if tissue processing induces a significant redistribution of calcium, it would not be possible to associate changes in calcium localization with a specific physiological treatment. To further define the usefulness of EM 45Ca ARG in studying calcium in smooth muscle, we have begun to examine the distribution of calcium under different physiological conditions. As part of a previous paper (McGuffee et al., 1985), we compared the distribution of developed 45Ca grains in the vas deferens of the guinea pig with that of the rabbit. Prior to freezing, the guinea pig tissue was rinsed for 6 min in nonradioactive physiological salt solution (PSS), whereas the rabbit tissue was not rinsed. These two tissues differed significantly in their relative calcium distribution in the cytoplasmic matrix, plasma membrane, and mitochondria. However, we could not determine whether the difference was due to species differences, to the rinse in nonradioactive PSS prior to freezing, or to both. In this paper we present the results of studies designed to sort out these differences. We also report the results of morphometric analyses of the vas deferens from the two species. MATERIALS AND METHODS Three male guinea pigs (300-400 gm) were sacrificed by a sharp blow to the head. The vasa deferentia were removed and immersed in a physiological salt solution (PSS) maintained a t room temperature. The PSS had the following composition in millimoles per liter: NaC1, 125; KCL2.7; CaC12, 1.8; glucose, 11; Tris (Trizma Base; Sigma Chemical Co., St. Louis, MO) buffer, 23.8. The Received February 25, 1986; accepted November 13,1986, Address reprint requests to Linda J. McGuffee, Ph.D., Department of Pharmacology, School of Medicine, University of New Mexico, Albuquerque, NM 87131. 322 L.J. McGUFFEE, S.A. LITTLE, AND B.J. SKIPPER solution was adjusted to pH 7.5 with 6N HC1 and was saturated with 100% oxygen. Tissues were dissected into approximately 1 mm2 pieces and placed in radioactive PSS that contained 45CaC12 (1 mCYml) (total CaCl2 = 1.8 mM) for a period of 65-85 min. At the end of the incubation period, tissues were rapidly frozen. Tissue from 3 rabbits was also prepared according to this protocol (McGuffee et al., 1985). Tissue from 3 different guinea pigs was prepared according to this protocol, except a 6-min rinse in nonradioactive PSS was added immediately prior to freezing (McGuffee et al., 1981). All tissues were quick frozen against a highly polished copper bar chilled in liquid nitrogen and dehydrated under vacuum a t low temperature in a glass freezedrying apparatus. The dried tissue was exposed to osmium tetroxide vapor in vacuo and then embedded in Spurr resin. Detailed descriptions of the freezing, freezedrying, and embedding procedures are given in previous publications (McGuffee et al., 1981; Chiovetti et al., 1987). A thin section through vas deferens cells that were quick frozen, freeze dried, and embedded is shown in Figure 1. Comparable tissue fixed in osmium tetroxide vapor for 2-3 hr, dehydrated in a graded series of ace- tones (McGuffee and Bagby, 1976), and embedded in Spurr resin is shown in Figure 2. Ice crystal damage is minimal in the freeze-dried preparation, and organelles and membranes are well preserved (compare Figs. 1,2). Eight blocks of tissue were selected, on the basis of their degree of morphological preservation, for electron microscopic examination. From each block, 100 nm sections were cut, placed on copper grids, and coated with a thin layer of carbon. The grids were then coated with a monolayer of Ilford L-4 emulsion. Following 8-10 wks exposure, autoradiograms were developed in Kodak D-19 developer, stained with uranyl acetate and Reynolds’ lead citrate, and examined in a Hitachi 11-C electron microscope. Micrographs were taken from grids on which the background grain density in tissue-free areas of plastic was 20% or less than the grain density over the tissue (McGuffee et al., 1985).A typical autoradiogram is shown in Figure 3. To determine the grain distribution, a circle of 275 nm radius is drawn around each grain (McGuffee et al., 1985). If a n organelle or plasma membrane is included within the circle, the grain is associated with that compound site. For example, the circled grain in Figure 3 is assigned to the compound site “plasma membrane + Fig. 1. Smooth muscle of the rabbit vas deferens, quick frozen, freeze dried, exposed to osmium vapor, and embedded in Spurr low viscosity resin. In this tangential section, part of a nucleus (n) can be seen in one cell. Extremely electron dense mitochondria (m) appear throughout the cells. Cells are separated from the extracellular space (e) by a plasma membrane (p). Surface vesicles (arrowheads) are seen along the membrane. Bar, 1pm. 45CALOCALIZATION IN VAS DEFERENS 323 Fig. 2. Smooth muscle cells of the rabbit vas deferens prepared using conventional techniques, i.e., fixed in osmium tetroxide vapor, dehydrated in a graded series of acetones, and embedded in S p u n low viscosity resin. The cells are separated from one another by extracellular space (e). Clusters of surface vesicles (arrowheads) are evident at discrete sites along the plasma membrane (p). Within the cell, mitochondria (m) and clear, membrane-bound spaces, presumably sarcoplasmic reticulum (s), are observed. Bar, 1 pm. cytoplasmic matrix + extracellular space.” The next step is to estimate the relative area of each component of the compound site. Using the above example, the area of the circle occupied by membrane, matrix, and extracellular space is determined from the morphometric analysis. From grain-counting and morphometric data, we calculate the grain density of the matrix farther than 275 nm away from the plasma membrane and organelles. Likewise, we calculate the grain density of the extracellular space away from the plasma membrane. If’ we assume that the grain density of the matrix inside the circle is the same as the grain density in the matrix away from the circle, we can estimate what proportion of the grains observed close to the membrane originated from the membrane and what proportion originated from the matrix. A similar determination is made for the extracellular space. The relative activity of the membrane can then be determined. Statistical Analysis Because of the relatively high energy (mean energy = 0.07 Mev; maximum energy = 0.25 Mev) of the beta particle emitted during 45Ca decay, a developed grain will frequently be observed outside of its source. This is especially true when the source of radiation is small relative to the range of the beta emission a s in the case of the plasma membrane, sarcoplasmic reticulum (SR), and mitochondria. Therefore, it is essential that any analysis used to evaluate autoradiograms of smooth muscle take radiation spread into consideration. Two types of information must be obtained from the autoradiograms. First, the distribution of 45Ca grains must be determined and, second, a morphometric analysis of the tissue must be carried out (Salpeter and McHenry, 1973; Williams, 1969; Williams, 1977). From these data the relative activity of each site (i.e., proportion of grains associated with the sitelproportion of area occupied by the site) can be determined. The rationale and development of the analysis has been described in detail elsewhere (McGuffee et al., 1981; McGuffee et al., 1985; Skipper and McGuffee, 1985). The following example illustrates what relative activity means in terms of our results. Assume that a cellular site occupies 50% of the area of the tissue. If 50% of the grains originate from the site, the relative activity is 1. If 100% of the grains originate from the site, the relative activity is 2. On the other hand, if a site occupies 1%of the total area of the tissue, and 5% of the grains are associated with the site, the relative activity is 5. If 50% of the grains are associated with the site, the relative activity is 50. Therefore, we can use relative activity to answer the question “How is calcium distributed among the cellular sites?” We cannot, however, use relative activity to answer the question “What is the millimolar concentration of calcium in a cellular site?” 324 L.J. McGUFFEE, S.A. LITTLE, AND B.J. SKIPPER Fig. 3. Smooth muscle of the guinea pig vas deferens incubated in 45Ca, rapidly frozen, freeze dried, osmicated, and embedded in Spurr low viscosity resin. A portion of a nucleus (n) is seen in one cell. Extremely electron dense mitochondria (m) are located both peripher- ally and centrally in the cell. Electron-translucent SR (s) has a similar distribution. Cells are separated by extracellular space (e).The relative activity of the organelles, plasma membrane, cytoplasmic matrix, and extracellular space is determined as described in the text. Bar, 1 pm, bution of calcium during EM preparation will mask any physiologically induced changes in calcium distribution. If, on the other hand, we can associate the localization of calcium with a physiological stimulus, this would provide support for the usefulness of the EM ARG technique for studying calcium distribution in smooth muscle. In the nonrinsed guinea pig vas deferens (Table 1, first column of data), the cytoplasmic matrix and the mitochondrial distributions are random (i.e., not different from 1.00 [P > .05]). In contrast, there are significantly more grains associated with the SR and the plasma membrane than would be expected if the distribution were random (P <.001). The nucleus contains less calcium than would be expected from a random distribution (P < .05). Thus, there is a selective association of calcium with some (but not all) cellular sites; the sites of significant accumulation, the plasma membrane and SR, agree with biochemical data showing calcium uptake into subcellular fractions containing these organelles (Carsten and Miller, 1980; Daniel et al., 1983; Raeymaekers et al., 1983). RESULTS In experiments in which the effect of rinsing in nonThe first question we had to address before we could radioactive calcium prior to freezing was investigated, interpret our results with confidence was, “Does the EM we observed that most of the radioactive calcium is ARG technique produce significant translocation of cel- removed by a 6-min rinse. The average grain density of lular calcium?” If translocation is a problem, redistri- the tissue prior to rinsing was 8.6 grainsl.6 pm2. The Once the relative activity of the cellular sites is determined, the standard error can be estimated using the methods developed by Skipper and McGuffee (1985). From examining standard error estimates obtained according to these equations, it becomes obvious that the smaller the cellular site, the larger the standard error. The reason for this is that the standard error is inversely proportional to the proportion of the area of the circle occupied by that site (see equation 10 in Skipper and McGuffee, 1985). Thus, the standard error will be large for a site that occupies 1%of the area, and will be small for a site that occupies 90% of the area. For the morphometric analysis, a standard point counting method was used to estimate the relative (i.e., fractional) volume occupied by each of the cellular sites (Weibel and Bolender, 1973). A double lattice test grid with scale grid spacing of 275 nm was superimposed on a micrograph (McGuffee et al., 1985). The number of points falling on each cellular site was counted and summed from all the micrographs. 45CA LOCALIZATION IN VAS DEFERENS TABLE 1. Relative activities of cellular sites in nonrinsed 325 From the data we were also able to estimate the relaand rinsed guinea pig vas deferens tive volume of the cell occupied by organelles, the plasma Nonrinsed relative Rinsed relative' Z membrane, and the cytoplasmic matrix in the two speCellular site activitv f SE activitv + SE value2 cies (Table 3). Chi square analysis indicates that the two species differ in the volume fraction occupied by the Total grains 2,853 277 cellular constituents. In particular, the nuclear volume Cytoplasmic 0.98 f 0.02** 0.77 f 0.06 3.43 fraction is larger in the rabbit than in the guinea pig, matrix and accordingly the cytoplasmic matrix is greater in the Plasma 2.60 f 0.45** 11.30 f 1.90 -4.45 guinea pig than in the rabbit. membrane Mitochondria 0.31 f 0.66* 8.26 f 2.73 -2.83 DISCUSSION SR 3.87 f 0.46 4.38 f 1.87 -0.26 Nucleus 0.73 f 0.11 1.18 f 0.40 -1.06 We have previously published the results of 45Ca EM ARG studies that were carried out on freeze-dried vas 'Data from McGuffee et al., 1985. 2Significance levels for z value: deferens of the guinea pig and the rabbit (McGuffee et I Z 1 Q 1.96 then N.S. al., 1985). Major differences in calcium localization were 2.58 < 1 Z I ~ 3 . 2 ,001 9 < P < .01 observed, but, because the experimental conditions prior 3.29 < I Z I P < ,001. to quick freezing of the tissues were not identical in the *P < .01. **P < ,001. two studies, we were unable to fully evaluate the differences. Thus, we could only speculate as to whether the observed distributions were due to rinsing in nonraaverage grain density after rinsing was 2.11.6 pm2. This dioactive calcium, due to species differences, or due to represents a decrease in grain density after rinsing of both. The present study was designed to sort out these possibilities. about 75%. In interpreting studies that examine the effect of rinsThe next step was to determine whether the grain distribution was the same after rinsing as it was before. ing on calcium distribution, one must remember that One would expect that if grains are removed uniformly the concentration of calcium per se in the external mefrom sites of localization, the relative activity of each dium is the same in both the incubation and the rinse site will remain the same after rinsing, even though the solution. The difference between solutions is that radiototal grain density has decreased. On the other hand, if active calcium in the incubation solution has been rea site decreases in relative activity after rinsing, this would indicate that the rate of removal is nonuniform and that calcium in this site exchanges more rapidly TABLE 2. Relative activities of cellular sites in nonrinsed than does calcium in sites where there is no change in guinea Dip. and nonrinsed rabbit vas deferens relative activity. The converse is true if the relative Z Guinea pig relative Rabbit relative' activity of a site increases. activity f SE activity f SE value2 Grain distributions with and without rinsing in non- Cellular site radioactive calcium for 6 min are given in Table 1. The Total grains 2,853 1,180 0.98 f 0.02* 1.05 & 0.03 2.24 relative activities of the plasma membrane and the mi- Cytoplasmic tochondria were significantly higher after rinsing; the matrix 2.27 f 0.49 -0.05 2.60 f 0.45 cytoplasmic matrix was significantly less radioactive Plasma after rinsing; the SR and the nucleus were not signifi- membrane 0.31 f 0.66 2.05 k 0.90 1.55 cantly different after rinsing. This suggests that radio- Mitochondria 2.02 f 0.92 -1.81 3.87 f 0.46 SR active calcium in the cytoplasmic matrix exchanges most Nucleus 0.73 0.11 0.69 f 0.10 -0.31 rapidly with nonradioactive calcium, SR and nuclear calcium exchange less rapidly, and plasma membrane 'Data from McGuffee et al.. 1985. 'Significance levels for z value: and mitochondrial calcium exchange least rapidly. I Z I g 1.96 then N.S. Next, we compared the distribution of calcium in non- 1.96 < I Z I g 2.58 .01 < P < .05. rinsed vas deferens from the rabbit (McGuffee et al., *P < .05 1985) and the guinea pig. The results of the analysis are shown in Table 2. When plasma membrane from the TABLE 3. Volume of cellular sites as a percentage of total guinea pig was compared with plasma membrane from cellular volume in the vas deferens the rabbit, the relative activity was not different. SimiCellular site* Guinea pig % Rabbit% larly, the relative activity for mitochondria, SR, and the nucleus from guinea pig were not different from the Cytoplasmic 90 (6298)' 82 (3173) relative activity of the same organelle in the rabbit. This matrix suggests that under identical physiological conditions, a Plasma 3 (204) 4 (171) membrane given organelle or the plasma membrane has similar 1(83) 2 (68) calcium sequestration properties in both species. In con- Mitochondria 3 (212) 2 (73) trast, the distribution of calcium in the cytoplasmic ma- SR Nucleus 3 (241) 10 (381) trix was different in the two species. The matrix of the 100 (7038) 100 (3866) guinea pig had significantly less calcium associated with Tntal it than did the matrix of the rabbit. However, given the *Overall chi square = 231.67 (P < .001). absolute magnitude of this difference, the physiological 'Figures in parentheses indicate number of points falling over each cellular site. significance is questionable. 326 L.J. McGUFFEE, S.A. LITTLE, AND B.J. SKIPPER placed with nonradioactive calcium in the rinse solution. Thus, the overall number of 45Ca grains that we visualize with ARG should decrease after rinsing as nonradioactive calcium replaces radioactive calcium in the cell. This is, in fact, what we observed. We also observed a significantly different pattern of distribution of the 45Cathat remained in the tissue after rinsing. For example, rinsing caused a 21% decrease in the relative activity of the cytoplasmic matrix (Table 1). This suggests that there is a pool of calcium in the matrix that is accessible to extracellular calcium and that turns over more rapidly than does the calcium associated with other sites. Presumably the calcium that remains in the matrix after rinsing is bound to proteins. The identity of these proteins is unknown; however, by use of X-ray microanalysis, Bond et al. (1984b) have postulated the existence of high-affinity calcium-binding proteins in the matrix that may act a s calcium buffers in vascular smooth muscle. Our data suggest that similar binding proteins may exist in the vas deferens. The fate of the calcium that is removed by rinsing is undetermined. However, it is not unreasonable to assume that it moves from the matrix to the plasma membrane. From the plasma membrane it can be removed from the cell by calcium extrusion mechanisms (Daniel et al., 1983; Van Breemen et al., 1980). There is a selective association of calcium with the plasma membrane in all of our experiments. This is in agreement with studies that have shown calcium uptake into enriched membrane fractions (Daniel et al., 1983; Kwan et al., 1983; Raeymaekers et al., 1983). In addition, this selective association increases after rinsing. An increase in relative activity of the plasma membrane after rinsing may indicate that this pool of calcium exchanges slowly with the extracellular calcium. Alternatively, it may be a pool that exchanges readily with extracellular calcium, but is continuously replenished with radioactive calcium released from internal sites such as the cytoplasmic matrix. In either case, these data support the concept that the plasma membrane plays a major role in cell calcium regulation. Mitochondria, isolated from a variety of different smooth muscles, can accumulate calcium in a n energy dependent manner (Batra, 1982; Ford and Hess, 1975; Janis et al., 1977). However, mitochondrial uptake of calcium does not appear to play a major role in regulating cytoplasmic calcium a t concentrations below about 1pM (Daniel et al., 1983; Ford and Hess, 1975; Janis et al., 1977; Johansson and Somlyo, 1980). In our studies, the free internal calcium concentration should be well below 1 pm since the cells have not been exposed to a n agonist to induce contraction (Filo et al., 1965). Consequently, mitochondria are not likely to act as a n important calcium sink under these conditions. On the other hand, the increase in the relative activity of the mitochondria after rinsing may indicate the presence of a slowly exchanging pool of calcium. This explanation is consistent with our data and with the hypothesis that mitochondrial calcium is relatively inaccessible a t physiological calcium concentrations. As with the plasma membrane, there is a selective association of calcium with the SR in both the guinea pig and the rabbit. Unlike the plasma membrane, there was no significant difference in the relative activity of the SR before or after rinsing. This suggests that the SR does not sequester calcium as tightly as do the mitochondria, nor does it turnover a s rapidly as does matrix calcium. The SR seems to function as a n intermediately exchangeable calcium pool. This observation is consistent with the idea that SR calcium may play a role in contraction (Bond et al., 1984a,b;Johansson and Somlyo, 1980). The nucleus does not accumulate calcium, nor does rinsing significantly alter the relative activity of this organelle in the guinea pig. Furthermore, the guinea pig nuclear relative activity is not significantly different from that of the rabbit. The failure of nuclei to sequester calcium has also been shown in vascular smooth muscle (Somlyo et al., 1979). Morphometric analysis indicates that the nuclear volume in the rabbit is less than in the guinea pig. It follows then that the relative volume of some other component of the cell must decrease to accommodate this increase. In the guinea pig vas deferens, this is accomplished via a reduction in the volume occupied by the matrix and not by changes in the relative volume of the organelles. Further study is required to determine whether there is a physiological significance to the increase in nuclear volume fraction. When tissues from the rabbit and the guinea pig were quick frozen without rinsing, the distribution of calcium was quite similar in the two species (Table 2). The relative activities of organelles and plasma membrane were not significantly different between species. The matrices in the two species were slightly different, but given the small magnitude of this difference, physiological relevance is doubtful. In summary, we have demonstrated that the relative distribution of calcium in the cell can be altered by physiological means. Thus, our results are not consistent with significant translocation of calcium during tissue processing. The data indicate that the same cellular sites that showed significant changes when we compared the vas deferens of the rabbit (nonrinsed) and the guinea pig (rinsed) are also significantly different when we compare the nonrinsed and rinsed guinea pig vas deferens. This indicates that species does not play a major role in determining the observed distribution of calcium in vas deferens from rabbit and guinea pig. The relative distributions before and after rinsing show several significant differences, indicating that the rate of calcium turnover is not uniform among the cytoplasmic matrix, plasma membrane, mitochondria, SR, and nucleus. ACKNOWLEDGMENTS This work was supported by National Institutes of Health grant 1-R01-GM30003. L.J.M.’s work was done during the tenure of a n Established Investigatorship from the American Heart Association and with funds contributed in part by the New Mexico affiliate. LITERATURE CITED Batra, S. (1982) Some aspects of calcium uptake by human myometrial mitochondria and microsomes relevant to relaxation. Acta Physiol. Scand., 114:91-95. Bond, M., T.Kitazawa, A.P. Somlyo, and A.V. Somlyo (1984a) Release and recycling of calcium by sarcoplasmic reticulum in guinea-pig portal vein smooth muscle. J. Physiol. (Lond.), 355t677-695. 45CA LOCALIZATION IN VAS DEFERENS Bond, M., H. Shuman, A.P. Somlyo, and A.V. Somlyo (1984b) Total cytoplasmic calcium in relaxed and maximally contracted rabbit portal vein smooth muscle. J. Physiol. (Lond.), 357t185-201. Casten, M.E., and J.D. Miller (1980)Characterization of cell membrane and sarcoplasmic reticulum from bovine uterine smooth muscle. Arch. Biochem. Biophys., 204~404-412. Chiovetti, R., L.J. McGuffee, S.A. Little, E. Wheeler-Clark, and J. Brass-Dale (1987) Combined quick-freezing, freeze-drying, and embedding tisue at low temperature and in low viscosity resins. J. Electron Microsc. Tech. 5tl-15. Daniel, E.E., A.K. Grover, and C.Y. Kwan (1983)Calcium. In: Biochemistry of Smooth Muscle, Vol. 111. N.L. Stephens, ed. CRC Press, Inc., Boca Raton, FI., pp. 1-80. Filo, R.S., D.F. Bohr, and J.C. Ruegg (1965) Glycerinated skeletal and smooth muscle: Calcium and magnesium dependence. Science, 1471581-1583. Ford, G.D., and M.L. Hess (1975)Ca-accumulatingproperties of subcellular fractions of bovine vascular smooth muscle. Circ. Res., 37.580587. Hudgins, P.M., and G.B. Weiss (1968) Differential effects of calcium removal upon vascular smooth muscle contraction induced by norepinephrine, histamine, and potassium. J. Pharmacol. Exp. Ther., 159t91-97. Janis, R.A., D.J. Crankshaw, and E.E. Daniel (1977) Control of intracellular Ca2+ activity in rat myometrium. Am. J. Physiol., 232:C50-C58. Johansson, B., and A.P. Somlyo (1980) Electrophysiology and excitationcontraction coupling. In: Handbook of Physiology, Section 2, The Cardiovascular System. D.F. Bohr, A.P. Somlyo, and H.V. Sparks, eds. Waverly Press, Baltimore, pp. 301-323. Kwan, C.Y., R.M.K.W. Lee, and A.K. Grover (1983) Subcellular membrane fractionation of rat vas deferens smooth muscle. Mol. Physiol., 353-69. McGuffee, L.J., and R.M. Bagby (1976) Ultrastructure, calcium accumulation, and contractile response in smooth muscle. Am. J. Physiol., 230: 1217-1224. McGuffee, L.J., S.A. Little, and L. Hurwitz (1979) Ultrastructural characteristics of freeze-dried smooth muscle. J. Histochem. Cytochem., 27~1267-1270. McGuffee, L.J., L. Hurwitz, S.A. Little, and B.E. Skipper (1981)A 45Ca autoradiographic and stereological study of freeze-dried smooth 327 muscle of the guinea pig vas deferens. J. Cell Biol., 90:201-210. McGuffee, L.J., E.S. Wheeler-Clark, B.J. Skipper, and S.A. Little (1985) The cellular distribution of calcium in freeze-dried rabbit vas deferens using EM autoradiography. Anat. Rec., 211:117-125. McGuffee, L.J., and S.A. Little (1986) Effect of osmium vapor on calcium loss during sectioning of freeze-dried, embedded tissue. Proc. Electron Microsc. Soc. Amer., 44:258-259. Popescu, L.M., and I. Diculescu (1975) Calcium in smooth muscle sarcoplasmic reticulum in situ. J. Cell Biol., 67:911-918. Raeymaekers, L., F. Wuytack, J. Eggermont, G. De Schutter, and R. Casteels (1983)Isolation of a plasma-membrane fraction from gastric smooth muscle. Biochem. J., 210:315-322. Salpeter, M.M., and F.A. McHenry (1973) Electron microscope autoradiography. In: Advanced Techniques in Biological Electron Microscopy. J.K. Koehler ed., Springer-Verlag, Berlin, pp. 113-152. Skipper, B.J., and L.J. McGuffee (1985) Statistical significance tests for autoradiographic data. Anat. Rec., 211t126-132. Somlyo, A.P., A.V. Somlyo, and H. Shuman (1979) Electron probe analysis of vascular smooth muscle: composition of mitochondria, nuclei, and cytoplasm. J. Cell Biol., 81r316-335. Sugi, H., and T. Daimon (1977) Translocation of intracellularly stored calcium during the contraction-relaxationcycle in guinea pig taenia coli. Nature (Lond.),269t436-438. Suzuki, S. (1982) Physiological and cytochemical studies on activator calcium in contraction by smooth muscle of a sea cucumber, IsastC chopus badionotus. Cell Tissue Res., 222t11-24. Van Breemen, C., P. Aaronsen, R. Loutzenhiser, and K. Meisheri (1980) Calcium, calcium antagonists, and cardiovascular disease. Chest, 78t157-165. Weibel, E.R., and R.P. Bolender (19731 Stereological techniques for electron microscopic morphometry. In: Principles and Techniques of Electron Microscopy, Vol. 3. M.A. Hayat, ed., Van Nostrand Reinhold, New York, pp. 237-296. Williams, M.A. (1969)The assessment of electron microscopic autoradiographs. In: Advances in Optical and Electron Microscopy, Vol. 3. R. Barer and V.E. Cosslett, eds., Academic Press, New York, pp. 219-277. Williams, M.A. (1977) Quantitative methods in biology. In: Practical Methods in Electron Microscopy, Vol. 6. A.M. Glauert, ed. North Holland Publishing Co., Amsterdam, pp. 85-169.