In vivo demonstration by radioautography of binding sites for insulin in liver kidney and calcified tissues of the rat.код для вставкиСкачать
THE ANATOMICAL RECORD 214:130-140 (1986) In Vivo Demonstration by Radioautography of Binding Sites for insulin in Liver, Kidney, and Calcified Tissues of the Rat B. MARTINEAU-DOIZE, M.D. McKEE, H. WARSHAWSKY, AND J.J.M. BERGERON Department of Anatomy, McGill University, Montreal, Quebec, Canada ABSTRACT An in vivo binding assay using radioautography was employed to visualize insulin receptors in rat tissues. Two and one-half minutes after the intravenous injection of 1251-insulin,free hormone was separated from bound hormone by whole body perfusion with lactated Ringer’s solution followed by perfusion with glutaraldehyde. The localization of bound hormone, fixed in situ by perfusion with glutaraldehyde, was determined. Nonspecific binding of labeled insulin was noted in the proximal convoluted tubules of the kidney cortex, prebone and adjacent bone, predentin and adjacent dentin, and enamel. Specific binding sites were observed a t the periphery of hepatocytes, over osteoblasts, and in relation to the endothelial cells of fenestrated capillaries within the papillary layer of the maturation zone of the incisors. In vivo radioautographic assays have demonstrated the presence of specific insulin receptors in many tissues (Bergeron et al., l977,1980a,b; van Houten et al., 1979). However, only one in vivo study has shown specific receptors for insulin in endothelial cells (van Houten and Posner, 1979). On the other hand, several investigators have identified specific receptors in cultured human and bovine endothelial cells originating from macrovessels (Peacock et al., 1982; Kaiser et al., 1982a,b; Carson et al., 1983) and also from microvessels (Pillion et al., 1982; King et al., 19831, as well as from microvessels of intact rat heart (Bar et al., 1982). In the present study, nonspecific insulin binding sites were observed in vivo in the proximal convoluted tubules of the kidney, the collagenous matrices of prebone and bone, predentin and dentin, and in the enamel of the rat incisor. Specific binding sites were seen at the periphery of hepatocytes, osteoblasts, and endothelial cells of fenestrated capillaries in the enamel organ of the rat incisor, but only in the zone of maturation. MATERIALS AND METHODS Animal Procedures Three series of experiments were carried out on 12 male Sherman rats weighing 95 +_ 8 gm (Tables 1, 2). Under pentobarbital anesthesia, the experimental rats were injected via the external jugular vein with 0.2 ml of freshly prepared 1251-insulin(specific activity, respectively, of 155 pCi/pg, 135 pcilpg, and 148 pCi/pg; Table 2). The iodination of porcine insulin (Porcine zinc insulin, Connaught Laboratories, Willowdale, Ontario) was performed by the chloramine T method as previously described (Posner et al., 1978). Precisely 2 minutes and 30 seconds after the injection, the rats were perfused through the left ventricle with lactated Ringer’s solution until the liver had visibly blanched (20-30 seconds). This washing, presumably to remove the free hormone 0 1986 ALAN R. LISS, INC. from the whole body, was followed immediately by perfusion for 10 minutes with 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer containing 0.05% CaC12 (pH 7.3). Control animals received the same quantity of 1251insulin plus 50 or 100 pg of unlabeled insulin (Table 2). Immediately after perfusion, kidney, liver, and tibia samples, as well as the mandibular incisors, were taken and immersed for 2 additional hours in the same fixative a t 4°C. In addition, portions of liver and kidney were weighed on a n analytical balance and examined for radioactive content (Table 1)using a Packard auto-a spectrometer (efficiency 41.5%; Packard, Dowmers and Grave, L). The noncalcified tissues were washed in 0.1 M sodium cacodylate buffer containing 0.05% CaC12, pH 7.3, and postfixed in 2% aqueous osmium tetroxide. After dehydration in graded concentrations of acetone, the tissues were embedded in Epon 812. The calcified tissues were decalcified in 4.13% disodium EDTA, pH 7.3, for 16 days a t 4°C (Warshawsky and Moore, 1967) and then washed for 24 hours in the same buffer, before being processed as above. Quantitative Analysis For light microscope radioautography, twelve l-pmthick sections of every block were cut with glass knives and placed in rows of four on each of three slides. The sections were prestained with iron hematoxylin, dipped in Kodak NTB2 liquid emulsion (Kopriwa and Leblond, 19621, and developed in three sets: one set was exposed for 1week, another set for 3 weeks, and the last set for Received October 1, 1984; accepted September 6, 1985. Address reprint requests to Dr. H. Warshawsky, Department of Anatomy, McGill University, 3640 University Street, Montreal, Quebec, H3A 2B2 Canada. 131 INSULIN BINDING IN RAT TISSUES TABLE 1. Content of radioactivity present in fixed liver and kidney tissues after injection of ‘2511-insulin Experiment No. 1A Experimental Control 1B Experimental Control 2A Experimental Control 2B Experimental Control 3A Experimental Control 3B Experimental Control 1251-insu~in injected (dpm)’ 422 x 106 422 x lo6 422 x 106 422 x lo6 432 x 432 x 432 x 432 x lo6 lo6 lo6 lo6 lo6 642 x 642 x lo6 642 x 642 x lo6 lo6 Concentration of radioactivity in fixed liver (dpdgwt)’ lo6 + 10.17 x lo6 lo6 1.03 X lo6 25.07 x lo6 f 12.06 x lo6 6.89 x lo6 f 0.62 x lo6 27.75 x lo6 & 2.21 x lo6 3.59 x lo6 f 0.51 x lo6 34.04 x lo6 k 15.80 x lo6 3.53 x lo6 & 0.50 x lo6 26.63 x lo6 f 3.52 x lo6 7.36 x lo6 k 0.43 x lo6 34.47 x lo6 & 8.41 x lo6 6.33 x lo6 k 0.35 x lo6 28.36 x 8.02 X P3 <’03’ Concentration of radioactivity in fixed kidney (d~m/gwt)~ 75.39 x 106 106.47 x lo6 <‘04’ 62.31 x 83.64 x <.Ooo 62.31 x 112.81 x <.‘I4 60.22 x 108.50 x <.Ooo 36.28 x 133.87 x <.002 p3.4 121.51 x 142.81 X lo6 lo6 lo6 k 13.78 x lo6 lo6 f 37.26 x lo6 lo6 f 10.32 x lo6 lo6 k 31.49 x lo6 lo6 k 9.29 x lo6 lo6 k 31.29 x lo6 lo6 f 25.85 x lo6 lo6 31.22 X lo6 <’055 <.06’ <.Oo2 <‘187 ‘Disintegrations per minute. 2Disintegrations per minute per gram wet weight of tissue. 3By Student’s t-test. 4The experimental values never exceeded the control. TABLE 2. Specific insulin-binding sites in the papillary layer of the rat mandibular incisor as assessed by quantitative light microscope radioautography Experiment No. 1A Experimental Control 1B Experimental Control 2A Experimental Control 2B Experimental Control 3A Experimental Control 3B Experimental Control Animal (g) * 1’5~-~nsulin SA (pCilpg) “Hot” insulin’ “Cold” insulin (PLg) (Pd Grain concentration’ (grains/3,450 pm’ mean f SD) p3 87 90 155 155 100 100 50 355.07 i 50.87 215.13 i 60.24 < .001* 88 85 155 155 100 100 50 348.88 i 63.96 217.86 + 39.79 < ,001” 90 92 135 135 100 100 50 375.83 k 67.46 201.42 35.60 < .001* 90 90 135 135 100 100 50 522.35 f 128.49 319.00 i 53.98 < .001* 103 105 148 148 100 100 100 641.16 f 129.14 215.15 29.96 < ,001” 103 103 148 148 100 100 100 528.20 i 63.05 260.40 i 44.37 < ,001“ ’“Hot” insulin, lZ51-labeledinsulin; “cold” insulin, nonradioactive insulin. 20ver 20,000 grains were counted for each experiment at a magnification of x 1,000. 3By Student’s t-test. *The experimental value exceeded the control by factors of 1.6 and 1.6 in experiment 1, 1.9 and 1.6 in experiment 2, and 2.9 and 2.0 in experiment 3. 6 or 8 weeks. Coverslips were mounted in Epon, which were 39, 81, and 112 days, respectively, for the three experiments. was Dolvmerized overnight at 65°C. F& efectron microscope radioautography silver-to-gold Analysis of Electron Microscope Radioautographs interference color sections were cut with a diamond knife The radioautographs from the papillary layer of the and prepared according to Kopriwa (1973). Fine-grain development was carried out by the Agfa-Gevaert 30- incisor maturation zone were photographed without bias lution-physical” technique as described by Kopriwa wherever silver grains were seen. The initial magnifi(1975). The exposure times for these radioautographs cation was 16,000x and the negatives were enlarged to 132 B. MARTINEAU-DOIZE, M.D. McKEE, H. WARSHAWSKY, AND J.J.M. BERGERON 50,000x magnification. A total of 542 silver grains were quantitated by direct scoring of the structures underlying the grains (Bergeron et al., 1977).The silver grains consisted either of single, compact spherical silver deposits or small clusters of two or more silver deposits. A cluster of silver deposits was interpreted as one silver grain when it fit within the space of a silver bromide crystal (Kopriwa, 1975). To assign a cluster to one silver grain and to determine its center, a transparency with a 7.0-mm circle (corresponding to the mean 140-nm diameter of a silver bromide crystal a t 50,OOOx magnification) was placed over the grain. The shortest distance from the geometrical center of each silver grain to the luminal endothelial cell membrane and to the abluminal membrane adjacent to the extracellular space outside the capillary in the papillary layer was measured with a Bausch and Lomb measuring magnifier. RESULTS Content of Radioactivity Present in Fixed Liver and Kidney Tissues TABLE 3. Grain counts over osteoblasts in the proximal metaphysis of the tibia after injection of '251-insulin Experiment No. 1A Experimental Control 1B Experimental Control 2A Experimental Control 2B Experimental Control Light Microscope Radioautography In the experimental rats, most silver grains were observed over the periphery of the hepatocytes and the adjacent sinusoids. The control rats showed numerous grains over the sinusoids, while only a few grains were localized over the hepatocytes. This grain distribution was similar to that described previously (Bergeron et al., 1977). Kidney An intense radioautographic reaction was present over the brush border of the proximal convoluted tubules from both experimental and control rats. The reaction intensity over cross-sectioned tubules was about the same in both experimental and control animals. These reactions thus represented nonspecific binding of labeled insulin to high-capacity sites and probably were related to the resorption and degradation of the hormone from the urinary filtrate (Bergeron et al, 1980a). Bone A moderate radioautographic reaction was found over the tibia1 bone of both experimental and control animals (Figs. 1, 2). The silver grains were located over the prebone and the adjacent bone covering the calcified cartilage of the mixed spicules in the proximal metaphysis of the tibia. Since the concomitant administration of a n excess of unlabeled insulin together with the labeled insulin did not produce a competitive inhibition of the bone labeling, it was concluded that the reaction represented nospecific binding to high-capacity sites. Silver grains also were present over the osteoblasts from experimental rats. Grain counts over the osteoblasts in experimental rats exceeded the counts in control rats by factors of 1.5 to 1.8. Therefore, these osteoblast reactions represented binding of labeled insulin to specific sites (Figs. 1,2, Table 3). Chondrocytes, Grain concentration (grains1344 pm2 f SD) P 8 8 85.05 f 23.15 48.50 & 15.32 <.ool 8 8 78.40 f 18.54 44.71 f 11.60 <.ool 6 6 54.31 f 13.00 34.75 f 9.85 <.ool 6 6 44.81 29.43 + 9.80 + 9.60 < .001 TABLE 4. Number of grains' over the predentin of the mandibular incisor of rats iniected with '251-insulin The radioactivity in experimental and control rats was assessed in fixed liver and kidney tissue using the Pack- Experiment ard auto-a spectrometer. Significant reduction of radio- No. activity was observed in the liver samples from control rats, while the radioactivity in control kidney samples 1A Experimental increased (Table 1). Liver Exposure time (weeks) Control 1B Experimental Control 2A Experimental Control 2B Experimental Control 3A Experimental Control 3B Experimental Control Grain concentration (grains/69O Fm2 mean + SD) P2 197.8 & 25.6 243.0 f 38.8 < .033* 168.0 f 38.8 217.2 -1: 20.7 < .0017* 115.1 + 20.6 171.5 + 27.4 < .0001* 153.0 f 32.2 232.1 f 41.8 < .0001* 243.4 + 26.3 271.5 f 33.7 < .0261* 192.8 f 32.9 249.4 + 42.0 < .0018* 'Counted in light microscope radioautographs exposed for 21 days over 6,900 pm' of predentin on the labial surface of the dentin. 'By Student's t-test. *The experimental values never exceeded the control. osteocytes, and the older bone matrix from experimental as well as from control rats were not labeled. Dentin Numerous silver grains were observed over the predentin and adjacent dentin of the incisors of both experimental and control rats (Figs. 3, 4).Grain counts over the predentin demonstrated that the experimental animals always showed a lower count than the control animals (Table 4).Since no decrease was observed in the number of grains over the predentin of the control animals, it was concluded that the reaction represented nonspecific binding of the labeled hormone. Maturation Zone of the Incisor Enamel Organ: General Description of the Maturation Zone At the light microscope level, the maturation zone of the incisor enamel organ was characterized by a full thickness of enamel covered by alternating bands of INSULIN BINDING IN RAT TISSUES Figs. 1, 2. Zone of mixed spicules from the proximal metaphysis of the tibia of an experimental (Fig. 11 and a control animal (Fig. 2). Light microscope radioautographs of 1-pm-thickEpon sections exposed for 3 weeks. x600. Many silver grains are present over the unstained prebone and the darkly stained bone of the experimental rat (Fig. 1, solid arrow) and the control (Fig. 2, solid arrow).The periphery of osteoblasts are labeled in the experimental animal (Fig. 1, open arrows) but are unlabeled in the control (Fig. 2, open arrows). c, capillary; cc, calcified cartilage; ob, osteoblast. 133 Figs. 3, 4. Dentin (D), predentin (Pd), and odontoblasts (Od) of the lower incisor from an experimental (Fig. 3) and a control animal (Fig. 4). Light microscope radioautographs of 1-pm-thick Epon sections exposed for 3 weeks. ~ 6 0 0Numerous . silver grains are present over the predentin and the adjacent dentin from both experimental and control animals. 134 B. MARTINEAU-DOIZE, M.D. McKEE, H. WARSHAWSKY, AND J.J.M. BERGERON En Figs. 5, 6 . Enamel organ i n the maturation zone of the lower incisor from an experimental (Fig. 5) and a control animal (Fig. 6). Light microscope radioautographs of 1-Km-thick Epon sections exposed for 3 weeks. X520. In Figure 5, numerous silver grains overlie the periphery of the capillaries (c) of the papillary layer (PL). A weaker reaction is present over the cytoplasm of the ruffle-ended ameloblasts (ram). In Figure 6, the number of silver grains associated with the capillaries is reduced considerably, whereas i t remains the same over the ameloblasts. En, enamel space. ruffle-ended and of smooth-ended ameloblasts (Josephsen and Fejerskov, 1977). The proximal ends of the ameloblasts were in contact with the papillary layer cells which formed furrows perpendicular to the long axis of the incisor. Within the furrows capillaries were distributed at approximately two levels (Figs. 5,6). Under the electron microscope (Fig. 7), the vessels appeared as typical fenestrated capillaries (Garant and Gillespie, 1969). The narrow portion of the endothelial wall frequently was pierced with fenestrae, closed by diaphragms. Tight junctions connecting the neighbouring endothelial cells often were associated with marginal folds. A continuous lamina densa separated the endothelial cells from the narrow connective tissue space between the capillaries and the continuous lamina densa over the papillary layer cells. The papillary layer cells were separated from each other by intercellular spaces filled with numerous microvilli. Neighbouring papillary layer cells were connected by desmosomes. Numerous coated vesicles, tubular structures, and mitochondria were observed in the cytoplasm of the papillary layer cells (Figs. 7, 8). Light Microscope Radioautography of the Papillary Layer A survey on the entire enamel organ revealed a strong radioautographic reaction only over the papillary layer of the maturation zone from the incisors of experimental rats (Fig. 5). The silver grains seemed to be located at the periphery of the capillaries (Fig. 5). Radioautographs of the same regions from the control rats showed a much INSULIN BINDING IN RAT TISSUES Fig. 7. Electron micrograph of a portion of a fenestrated capillary in the papillary layer of the rat incisor enamel organ. ~50,000.Thin diaphragms are stretched across the fenestrae (arrows). L, lumen; E, endothelial cell cytoplasm; P, papillary layer cell; LD, lamina densa; m, mitochondria; cv, coated vesicle; mv, microvilli. Inset Electron micrograph illustrating the presence of marginal folds and tight junctions. ~50,000. 135 136 B. MARTINEAU-DOIZE, M.D. McKEE, H. WARSHAWSKY,AND J.J.M. BERGERON Fig. 8. Electron microscope radioautograph of the papillary layer of the enamel organ from the lower incisor of a rat 2.5 minutes after the injection of lz51-insulin. X27,OOO. Silver grains overlie the thinwalled endothelial cell (E) of the capillary. luminal boundary (LB);pericapillary space boundary (PB); P, papillary layer cell. 137 INSULIN BINDING IN RAT TISSUES Lumen -2000 1600 1200 800 400 0 400 ' 800 1200 1600 2000 Distance from the luminal cell membrane ( O ) ( In nrn) Fig. 9. Distribution of silver grains on either side of the luminal cell membrane (0) of capillaries in the enamel organ from rats at 2.5 minutes after the injection of 1251-insulin.The dashed vertical line represents the abluminal surface of the endothelial cell. Most of the grains are close to the luminal cell membrane. weaker reaction (Fig. 6). Quantitative analysis of the grainsl690 ,urn2),only a weak reaction was observed over radioautographs indicated a significant decrease (1.5-to the smooth-ended cells (66 grains/690 pm2). On the other 2.9-fold) in total number of grains per 3,450 pm2 be- hand, while a weak reaction was seen over the enamel facing the ruffle-ended cells (68 grains/690 pm2), many tween the experimental and control animals (Table 2). grains were present over the enamel facing the smoothElectron Microscope Radioautography of the Papillary Layer ended ameloblasts (192 grains/690 prn2). Since the silver Electron micrographs of the papillary layer from the grain distribution and density were identical for both maturation zone in experimental rats suggested that the experimental and control animals and thus the remost grains were associated with endothelial cells of the actions did not represent specific binding sites for the capillaries (Fig. 8). Quantitative analysis of the electron hormone, the reactions did indicate the ability of molemicrographs by the direct scoring method showed that cules as large as insulin to enter the enamel organ and 65% of the silver grains were either over the endothelial the enamel. cytoplasm or over the capillary lumen close to the endoDISCUSSION thelial cell or over the pericapillary space. The in vivo radioautographic method, with labeled A more detailed analysis of grain distribution was obtained by measuring the shortest distance between biologically active peptide hormones, has been used to the center of each silver grain and the luminal (Fig. 9) recognize and quantitate receptors for several hormones and abluminal (Fig. 10) endothelial cell membranes. in a variety of target cells (Bergeron et al., l977,1980a,b, These grain density histograms demonstrated that the 1981, 1983; Bergeron and Posner, 1979; Warshawsky et majority of silver grains were situated over cytoplasmic al., 1980). This specific binding assay is based on the application of the law of mass action to a living animal. components close to the luminal cell membrane. Thus, in control animals the concomitant administraLight Microscope Radioautography of the Ameloblasts and tion of high concentrations of unlabeled insulin (excess) Enamel in the Maturation Zone together with 1251-labeled insulin produces a competiThe cells of the enamel organ in the presecretory and tion for binding on the specific saturable receptor sites. secretory zones, as well as the enamel in the secretory Experimental animals receive the same amount of lazone, were unlabeled. In the maturation zone, the beled insulin only. Specific binding is defined as the smooth-ended and the ruffle-ended ameloblasts as well difference in bound labeled hormone between experias the enamel layer facing these cells showed different mental and control animals. Three separate experiments with 1251-insulininvolvgrain distributions (Figs. 11, 12). Whereas numerous grains were found over the ruffle-ended ameloblasts (241 ing a totaI of six experimental and six control rats gave 138 B. MARTINEAU-DOIZE, M.D. McKEE, H. WARSHAWSKY, AND J.J.M. BERGERON W I I I I I I I Extracellular Space " 2000 1200 Is'oo 860 ' 400 0 400 800 I200 I600 2000 Distance from the abluminol cell membrane ( 0 )( i n nrn) Fig. 10. Distribution of silver grains on either side of the abluminal cell membrane (0) of papillary layer capillaries of the enamel organ from rats 2.5 minutes after the injection of 1251insulin. The dashed vertical line represents the position of the luminal cell membrane. Again, most of the grains lie closer to the luminal cell membrane. Figs. 11, 12. Ameloblasts and the adjacent enamel from the maturation zone of the lower incisor of two control rats. Light microscope radioautographs of 1-pm-thick Epon sections exposed for 3 weeks. x 520. The enamel (En) facing the ruffleended ameloblasts (ram, Fig. 11) is almost unlabeled, while many grains overlie the enamel facing the smooth-ended ameloblasts (Sam, Fig. 12). On the other hand, the smooth-ended ameloblasts show only a few silver grains, while the ruffle-ended ameloblasts are prominently labeled. INSULIN BINDING IN RAT TISSUES similar results. Competitive inhibition, indicating specific binding, was seen in relation to the hepatocytes, osteoblasts, and capillaries of the papillary layer in the maturation zone of the continuously growing incisor. In the latter case, only these capillaries were labeled and electron microscope radioautography confirmed that the silver grains were over the endothelial cells of these fenestrated capillaries. Nonspecific binding was observed within the proximal convoluted tubules of the kidney, the prebone and adjacent bone layer in the mixed spicules of the tibia, the incisor predentin and adjacent dentin, and the enamel. Specific Binding of Insulin to Hepatocytes and Osteoblasts Using the in vivo assay with radioautography, specific receptors for insulin have been found in the exocrine pancreas, the liver, the columnar epithelial cells of the intestinal tract, the adrenal cortex and medulla, and the circumventricular organs of the brain (Bergeron et al., 1977, 1980a,b; van Houten et al., 1979). In the present experiments the results obtained previously with liver were confirmed, thus strengthening the reliability of the labeling seen in other tissues of the same animals. Specific binding sites at or near the surface of osteoblasts were demonstrated by quantitation in the region of mixed spicules from the proximal metaphysis of the tibia (Table 3). Similar binding was also seen over osteoblasts from other sites in the tibia and in calvarial bone. The demonstration of specific receptors for insulin in bone secretory osteoblasts suggests a direct role for this hormone in bone formation. Specific Receptors for Insulin in Capillaries of the Enamel Organ Van Houten and Posner (1979) observed specific insulin binding sites in endothelial cells of blood vessels in the brain. This is the only in vivo demonstration of insulin receptors in vascular endothelium. However, in vitro studies have demonstrated the presence of specific insulin binding sites in cultured human and bovine endothelial cells from macro- and microvessels (Kaiser et al., 1982a,b; Peacock et al., 1982; Pillion et al., 1982; Carson et al., 1983; King et al., 1983; Soda and Tavassoli, 1983) as well as in the endothelium of microvessels from isolated but intact rat heart (Bar et al., 1982). The discrepancy between the in vitro results obtained from cultured endothelial cells and the in vivo radioautographic assay may be due to the lack of sensitivity of the radioautographic method in detecting very low levels of insulin binding (Bergeron et al., 1980a).The presence of specific insulin receptors in concentrations high enough to be detected indicates a high-density receptor site. Such a site was shown to be present on the luminal surface of fenestrated endothelial cells lining capillaries within the papillary projections of the enamel organ only in the maturation zone. Analysis of electron microscope radioautographs showed that at 2.5 minutes after the intravenous injection of 1251-insulin,most of the silver grains were located over components of the endothelial cells close to the luminal surface. Significance of Insulin Receptors in Capillary Endothelium of the Enamel Organ The enamel organ in the postnatal rat incisor remains an embryonic system throughout life because of the 139 continuous eruption of this tooth. Blood vessels never penetrate the basement membrane on the outer surface of the enamel organ, but the vessels are brought closer to the ameloblasts by invaginations of the outer dental epithelium. These invaginations are developed most strikingly in the maturation zone of the incisor, after the ameloblasts have completed enamel secretion. The epithelial tissues form high and regular ridges which alternate with the insulin-receptor-containing capillaries. In longitudinally sectioned incisors, these ridges appear as papillae; hence, this layer is termed the “papillary layer” of the enamel organ. Its function is completely unknown, but is appears to be associated with the ameloblast layer in processes believed to occur in the enamel to affect its maturation. These processes include the loss of proteins and water, and the additional growth of the hydroxyapatite crystals. Papillary layer cells are surrounded by abundant extracellular space and contain large numbers of coated pits and vesicles. This appearance is suggestive of a resorptive function. The direction of movement, either from the enamel or from the vessels toward the enamel, or in both directions, has never been ascertained. The papillary layer cells are separated from the insulin-receptor-containing capillaries by 1)the basement membrane of the enamel organ; 2) a narrow extracellular connective tissue space; and 3) the basement membrane of the fenestrated capillaries. Within this environment the role of insulin may be related to endothelial functions, to enamel organ functions, or more specifically,to the function of enamel maturation. Whatever its significance,the occurrence of these receptors in close proximity to the enamel organ in maturation is the first documentation of any hormonal receptors in tooth formation. Significance of the Nonspecific Binding Sites Four sites of nonspecific binding of the labeled insulin molecules were noted. The proximal convoluted tubules of the kidney cortex showed an intense, nondisplaceable reaction in the vicinity of the brush border. Such reactions have been described with various labeled peptide hormones (Bergeron et al., 1980a, 1983; Warshawsky et al., 1980) and represent the normal protein-clearing function of the proximal convoluted tubules. The ubiquitous occurrence of this reaction is used to monitor the efficacy of the intravenous administration of labeled peptide hormones for in vivo radioautographic assays. The occurrence of nonspecific binding in prebone and adjacent calcified bone and in the predentin and adjacent dentin and in the enamel of the incisors is an interesting observation concerning the penetration of relatively large protein molecules into these extracellular matrices. In each case an epithelium or epitheliallike layer of cells separates the matrices from the blood supply. Hence, in order for a molecule of insulin with a molecular weight of 5,700 daltons to enter these matrices it would have to pass very rapidly through the cellular layers. Once in the matrices insulin molecules appear to diffuse throughout the collagenous stroma of prebone and predentin, and between the maturing hydroxyapatite crystallites of bone, dentin, and enamel. Furthermore, they must be bound strongly enough to these matrices to survive histological processing. Since all these events occur within 2.5 minutes after injection, it appears that the cellular layer (osteoblastsin the case 140 B. MARTINEAU-DOIZE, M.D. McKEE, H. WARSHAWSKY. AND J.J.M. BERGERON of bone, odontoblasts in dentin and ameloblasts in enamel) do not form barriers to the free passage of the lZ5I-labeledprotein molecules. The results regarding the passage of insulin across the enamel organ are particularly noteworthy. The zone of maturation is characterized by the presence of two types of ameloblasts and their modulation from the ruffle-ended to the smoothended types at least five times during the maturation process (Josephsen and Fejerskov, 1977). The present results showed labeled insulin in the enamel related to the smooth-ended cells, but the relative absence of this labeled molecule from the cell layer. In contrast, ruffleended cells showed labeling associated with the cells, but almost no labeling over the enamel in contact with this cell type. These findings agree with those of Sasaki (19841, who used horseradish peroxidase, a protein of 40,000 molecular weight. Although junctional complexes are present a t both ends of the maturation ameloblasts, the proximal complexes of both cell types (closest to the vascular supply) are permeable to large protein molecules. However, only the distal junctional complexes of the smooth-ended ameloblasts allow this protein molecule to pass into the enamel. ACKNOWLEDGMENTS This work was supported by grants from the Medical Research Council of Canada. Dr. B. Martineau-Doize is a recipient of a Medical Research Council Fellowship. The authors are indebted to Dr. B.I. Posner, Department of Medicine, McGill University, for generously providing the iodinated insulin. LITERATURE CITED Bar, R.S., A. Derose, W.G. Owen, A. Sandra, and M.L. Peacock (1982) Insulin receptors in microvessels of the intact heart. A kinetic and morphometric demonstration. Diabetes, 31:(Suppl. 21173. Bergeron, J.J.M., G. Levine, R. Sikstrom, D.O'Shaughnessy, B. Kopriwa, N.J. Nadler, and B.I. Posner (1977) Polypeptide hormone binding sites in uiuo: Initial localization of 1251-labeledinsulin to hepatocyte plasmalemma as visualized by electron microscope radioautography. Proc. Natl. Acad. Sci. U.S.A., 745051-5055. Bergeron, J.J.M., and B.I. Posner (1979) In uiuo studies on the initial localization and fate of polypeptide hormone receptors by the technique of quantitative radioautography. J. Histochem. Cytochem., 27,1512-1513, Bergeron, J.J.M., R. Rachubinski, N. Searle, D. Borts, R. Sikstrom, and B.I. Posner (1980a) Polypeptide hormone receptors in uiuo: Demonstration of insulin binding to adrenal gland and gastrointestinal epithelium by quantitative radioautography. J. Histochem. Cytochem. 282324-835. Bergeron, J.J.M., R. Rachubinski, N. Searle, R. Sikstrom, D. Borts, P. Bastien, and B.I. Posner (1980b) Radioautographic visualization of in uiuo insulin binding to the exocrine pancreas. Endocrinology, 107:1069-1080. Bergeron, J.J.M., L. Resch, R. Rachubinski, B.A. Patel, and B.I. Posner (1983) Effect of colchicine on internalization of prolactin in female rat liver: An in uiuo radioautographic study. J. Cell Biol., 96,875886. Bergeron, J.J.M., S. Tchervenkov, M.F. Rouleau, M. Rosenblatt, and D. Goltzman (1981) In uiuo demonstration of receptors in rat liver to the amino-terminal region of parathyroid hormone. Endocrinology, 109:1552-1559. Carson, M.P., S.W. Peterson, M.E. Moynahan, and D. Shepro (1983) Binding, internalization and degradation of 1251-insulinby cultured bovine aortic endothelial cells: effects of serotonin. In Vitro, 19,833-840. Garant, P.R., and R. Gillespie (1969) The presence of fenestrated capillaries in the papillary layer of the enamel organ. Anat. Rec., 163:7 1-80. Josephsen, K., and Fejerskov, 0. (1977) Ameloblast modulation in the maturation zone of the rat incisor enamel organ. A light and electron microscopic study. J. Anat., 124:45-70. Kaiser, N., I. Voldavsky, A. Tur-Sinai, Z. Fuks, and E. Cerasi (1982a) Binding, internalization and degradation of insulin in vascular endothelial cells. Diabetes, 31,1077-1083. Kaiser N., I. Vlodavsky, A. Tur-Sinai, Z. Fuks, and E. Cerasi (1982b) Insulin binding and degradation in vascular endothelial cells: Modulation by cell growth and culture organization. Endocrinology, 113:228-234. King, G.I., S.M. Buzney, C.R. Kahn, N. Hetu, S. Buchwald, S.G. MacDonald, and L.I. Rand (1983) Differential responsiveness to insulin of endothelial and support cells from micro- and macrovessels. J. Clin. Invest., 71:974-979. Kopriwa, B.W. (1973) A reliable, standardized method for ultrastructural electron microscope radioautography. Histochemie, 37: 1-17. Kopriwa, B.M. (1975) A comparison of various procedures for fine grain development in electron microscope radioautography. Histochemistry, 44:201-224. Kopriwa, B.M., and C.P. Leblond (1962) Improvement in coating technique of radioautography. J. Histochem. Cytochem., 10:269-284. Peacock, M.L., R.S. Bar, and J. Goldsmith (1982)Interactions of insulin with bovine endothelium. Metabolism, 31.52-56. Pillion, D.J., J.F. Haskell, and E. Meczan (1982) Cerebral cortical microvessels: An insulin-sensitive tissue. Biochem. Biophys. Res. Commun., 1043386-692. Posner, B.I., Z. Josefsberg, and J.J.M. Bergeron (1978) Intracellular polypeptide hormone receptors: Characterization of insulin binding sites in Golgi fractions from the liver of female rats. J. Biol. Chem., 253:4067-4673. Sasaki, T. (1984) Morphology and function of maturation ameloblasts in kitten tooth germs. Z-Anat., 138:333-342. Soda, R., and M. Tavassoli (1983) Distribution of insulin receptors in liver cell suspensions using a minibead probe. Highest density is on endothelial cell. Exp. Cell Res., 145t389-395. van Houten, M., and B.I. Posner (1979) Insulin binds to brain vessels in uiuo. Nature, 282r623-625. van Houten, M., B.I. Posner, B.M. Kopriwa, and J.R. Brawer (1979) Insulin-binding sites in the rat brain: in uiuo localization to the circumventricular organs by quantitative radioautography. Endocrinology, 105:666-673. Warshawsky, H., D. Goltzman, M.F. Rouleau, and J.J.M. Bergeron (1980) Direct in uiuo demonstration by radioautography of specific binding sites for calcitonin in skeletal and renal tissues of the rat. J. Cell Biol., 85:682-694. Warshawsky, H., and G. Moore (1967) A technique for the fixation and decalcification of rat incisors for electron microscopy. J. Histochem. Cytochem., 15.542-549.