Quantitative analysis of zonulae occuldentes between oviductal epithelial cells at diestrous and estrous stages in the mouseFreeze-fracture study.код для вставкиСкачать
THE ANATOMICAL RECORD 206:257-266 (1983) Quantitative Analysis of Zonulae Occludentes Between Oviductal Epithelial Cells at Diestrous and Estrous Stages in the Mouse: Freeze-Fracture Study KIYOTAKA TOSHIMORI, RUIKO HIGASHI, AND CHIKAYOSHI GURA Department ofAnatom y, Medical College of Myazakz, Mzyazakl, Japan 889-16 ABSTRACT The zonulae occludentes between oviductal epithelial cells were quantitatively analyzed at diestrous and estrous stages in the mouse, using the freeze-fracture technique. Zonulae occludentes were predominantly anastomosing a t the diestrous stage, while they were predominantly parallel a t the estrous stage. The lowest mean value of junctional strands comprising the zonulae occludentes was 5.3 rt 1.6. Parallel-type zonulae occludentes had more strands than the anastomosing type. Secretory cells usually had more strands than ciliated cells. The shallowest mean depth occupied by junctional domain was 0.51 k 0.20 pm. The depth was usually somewhat greater in anastomosing-type zonulae occludentes than in the parallel type. It was also slightly greater in ciliated cells than in secretory cells. The depth was likely to be greater a t diestrous stage than a t the estrous stage. However, neither the number of strands nor the depth was significantly different between diestrous and estrous stages in homologous types of zonulae occludentes. On the basis of these results, the zonulae occludentes in oviductal epithelium are considered to be morphologically of a tight type a t any time period throughout the estrous cycle. The results of lanthanum tracer experiments suggest that the zonulae occludentes in the oviductal epithelium do not; always function as a barrier to the exogenous tracer. These morphological phenomena are discussed in relation to mouse fertilization in vivo. The mammalian oviduct is customarily subdivided into three regions, isthmus, ampulla, and fimbriated infundibulium. In the tract, secretory cells (nonciliated) and ciliated cells are the primary types. Secretory cells are more abundant than ciliated cells in the isthmus, while the proportion of secretory cells decreases gradually from the isthmus to the ampulla. Such regional variations of the lining epithelium are well known (Clyman, 1966; Hafez, 1973). The response of the mammalian oviducts to the ovarian hormones (estrogen and progesterone) is also well known (Clyman, 1966; Beier, 1974; West et al., 1976; Komatsu and Fujita, 1978; Verhage et al., 1979; Odor et al., 1980; Bareither and Verhage, 1981). These hormonal changes regulate the amount and composition of the oviductal luminal fluid (Urzua et al., 1970; Blandau, 1971; Beier, r 1983 ALAN R LISS, INC 1974). The distended oviductal lumen contains fluid, conspicuously in the ampulla, which is rapidly accumulated under the influence of estrogen at the estrous stage. Progesterone causes closure of the oviductal lumen and resorption of luminal fluid at the diestrous stage. No detailed report has been published on freeze-fracture images of the oviductal epithelium, although Komatsu et al. (1979) reported two types of zonulae occludentes between adjacent epithelial cells in the mouse; i.e., parallel (few anastomosing) and anastomosing types. However, they did not mention regional differences in the distribution of zonulae occludentes in the oviduct during the estrous cycle. Received December 9,1982; accepted March 15, 1983. 258 K. TOSHIMOKI, R. HIGASHI, AND c. ~ U R A In general, zonulae occludentes function as a diffusion barrier to separate the internal environment of the body from the external world (Farquhar and Palade, 1963; Reese and Karnovsky, 1967; Claude and Goodenough, 1973; Staehelin, 1974). We were, therefore, interested in examining how the zonulae occludentes between oviductal epithelial cells are organized a t the time of fertilization, and whether there is a difference in the structure of the zonulae occludentes between the diestrous and estrous stages. A preliminary report of this work has appeared elsewhere (Toshimori et al., 1982). (diestrous stage) of the same strain at room temperature. The various regions of the oviduct were selected and cut into small pieces. The tissues were then postfixed with 2% osmium tetroxide in the same buffer described above. After dehydration by ethanol as usual, they were embedded in Epon 812. The sections were cut by a glassknife-equipped LKB ultrotome and stained with uranyl acetate. The replicas and thin sections were examined by JEOL 100B or 200CX electron microscopes at a n accelerating voltage of 80 kV. Quantitation for Zonulae Occludentes Ten animals were used for control oviducts MATERIALS AND METHODS a t the diestrous stage and 30 for the isthmus Freeze-Fracture and ampulla at the estrous stage. Fifty zonMale and female ddY strain mice, 6-12 ulae occludentes analyzable on cleaved weeks in age, were housed in a controlled planes were selected from each of secretory environment (7 AM - 7 PM:light; 7 PM - 7 AM: and ciliated cells in the control oviducts, and dark) and provided with food and water ad the isthmus and ampulla at the estrous stage. libitum. The females were induced to ovulate Therefore, the total number of zonulae occluwith a n intraperitoneal injection of 2 IU of dentes examined was 300. The following pregnant mare’s serum (PMS),followed by 2 characteristics of zonulae occludentes were IU of human chorionic gonadotropin (HCG), analyzed: 1) the number of zonulae occlu48 hours later. Such doses usually induced dentes with parallel or anastomosing type 1-20 eggs in the ampulla of each oviduct as (classification of the junctional type is depreviously reported by one of the authors scribed in Results), 2) the number of junc(Toshimori, 1982). They were mated with the tional strands, and 3) the depth along the males and sacrificed by cervical dislocation lateral plasma membrane occupied by junca t 0.5-3 hours after plug formation (10-15 tional strands. Analyses were made only in hours after HCG injection). Reproductive or- micrographs where junctions were exposed gans were dissected out into 2.5% glutaral- from apical to basal edge. Measurements for dehyde adjusted to pH 7.2 by 0.1 M cacodylate the second and third characteristics were buffer a t room temperature. The oviduct was made of the average number of strands and clearly divided into two regions, isthmus and of the average depth of a n exposed junction, ampulla. As a control for the diestrous stage respectively. These analytical data were ex(control oviduct), female mice were used in pressed as mean standard deviation of the which about 60 hours had elapsed after HCG mean (SD). Differences were assessed by x2injection. At this time, the boundary between test for the first characteristic and by Stuisthmus and ampulla (I-Ajunction) could not dent’s t-test for the second and third characbe clearly identified by stereoscope, since the teristics. P < .05 was used as indicating a lumen of the oviduct had already been re- significant difference. duced. Therefore, the control oviduct conRESULTS tained either isthmus only or both isthmus and ampulla. They were dissected and fixed The word “control oviduct” is used only for in the same manner described above. These the diestrous stage, while the words “isthsamples were then cryoprotected by 30% mus” and “ampulla” are used only for the glycerin overnight. The replicas were made estrous stage in this report. in a FD-2A type fracturing device as previFreezeFracture ously described (Toshimori and Oura, 1982). The zonulae occludentes in mouse oviducLanthanum Infusion tal epithelium resemble those of other epiA fixative containing 2.5% glutaraldehyde thelia in the general arrangement of their and 2% lanthanum nitrate (pH 7.2 adjusted elements (more or less continuous bands by 0.1 M cacodylate buffer) was infused made up of ridges on the P-face and grooves through the thoracic aorta in two adult mice on the E-face). Ciliated cells are easily differ- * 259 ZONULAE OCCLUDENTES IN MOUSE OVIDUCT TABLE 1. Number of zonulae occludentes between adjacrnt cells at various regions’ 8-C Cell Control oviduct (diestrous) Isthmus (estrous) Ampulla (estrous) 1 I1 I I1 I I1 s-s c-c 19 15 35 8 26 12 - ___ S C 9 26 25 20 22 6 10 6 6 9 9 3 2 10 11 8 I 1 Total 40 60 69 31 61 33 100 100 100 ‘S= secretory (nonciliated) cell (S-S = between adjacent secretory cellsi; C = ciliated cell (C-C = hetween adjacent cillated cells); I = type I (parallel]zonula occludens; I1 = type I1 (anastornosing) zonula occludens. entiated from secretory (nonciliated) cells by the presence of the so-called “necklace” at the ciliary base (Figs. 2, 3, 5) as reported by Inoue and Hogg (1977) and Komatsu et al. (1979). The zonulae occludentes were classified into two cases: 1) zonulae occludentes between adjacent secretory cells (Fig. 1)or between adjacent ciliated cells (Fig. 2) and 2) zonulae occludentes between secretory and ciliated cells (Fig. 3). Further, each zonula occludens was classified into two types: parallel type (I) or anastomosing type (11). The criteria for classification were as follows: Type I zonula occludens had many strands running parallel to the luminal surface, whose branching points were very few (Figs. 4, 5). Type I1 zonula occludens had many strands anastomosing frequently, whose branching points were much greater in number than those of type I (Figs. 1-3). Abluminal strands in these junctions frequently displayed terminal loops and/or free-ends (Figs. 2-5) as reported by previous workers (Pitelka et al., 1973; Staehelin, 1974; Tice et al., 1975; Suzuki and Nagano, 1978). Such a configuration was recognized as a new junctionai formation area (iunction assembiv or extension) by previouslworkers (Tice et“ al., 1975; Suzuki and Nagano, 1978).Further, in this study, terminal loops and/or free-ends were observed in both types of zonulae occludentes in both types of cells (Figs. 2-5). Therefore, we determined a type of zonula occludens, preferentially based on the structure in the adluminal strands, excluding the abluminal strands with terminal loops-and/ or free-ends. Zonulae occludentes with a structural nature of both types, which were sometimes observed in various regions, were ruled out for purposes of quantitation. The number of type I or I1 zonulae occludentes The number of type I or I1 zonulae occludentes on each cell is presented in Table 1. As to the numerical MI ratio in adjacent secretory cells, the difference between the control oviduct and the isthmus was statistically significant (P < .001), while the difference between the control oviduct and the ampulla was not significant (.05 < P < .lo). As to the numerical I/II ratio in adjacent ciliated cells, the difference was statistically significant (P < .001) in either case-the control oviduct versus the isthmus, and the control oviduct versus the ampulla. Such a tendency was shown between secretory and ciliated cells as well. As to the numerical I/II ratio in total number, the difference between diestrous stage and estrous stage was statistically significant (P < .001) in either casethe control oviduct versus the isthmus, and the control oviduct versus the ampulla. Therefore, the zonulae occludentes at diestrous stage were predominantly anastomosing, while those a t estrous stage were predominantly parallel. Junctional morpholom -The number of strands and the depth occupied by junctional strands is presented in Table 2. The lowest mean value of junctional strands comprising the zonulae occludentes was 5.3 f 1.6 in type I1 zonula occludens in the ampulla. Comparing the diestrous stage t o the estrous stage, no significant differA b breuiations D E P Desmosome E-face P-face Figs. 1-3. Type I1 (anastornosing) zonulae occludentes from the control oviduct (diestrous stage). Fig. 1. Zonula occludens between secretory cells. Grooves, which are registered with strands, anastomose frequently on the E-face. The depth occupied by junctional strands is about 1 pm. A cluster of pits which correspond to gap junctional particles is displayed on the E-face (circles). x 47,000. Fig. 2. Zonula occludens between ciliated cells. Ciliated cells are differentiated from secretory cells by the presence of the so-called ciliary necklace (asterisks).Note some abluminal strands show free-ends (arrowheads), A rectangular-shape cluster of intramembraneous particles is demonstrated on the P-face (circle). A three-cell junction is observed between double arrowheads. x 39,000. Inset) High magnification of many rectangular-shape clusters of particles similar t o that in encircled area of Fig. 2. x 90,000. Fig. 3. Zonula occludens between secretory (S) and ciliated (C) cells. Terminal loops (arrowheads) are seen. Desmosomal patches are also seen. x 36,000. Fig. 4. Type I parallel 'Onula occludens between secretory cells from the isthmus a t estrous stage. Grooves run parallel to each other in the adluminal region (cornpare Fig. 4 to Figs. 1-3). Free-ends are seen (arrowheads). x 73,000. Fig. 5 . Type I zonula occludens between ciliated cells from the ampulla at estrous stage. Some abluminal strands also show free-ends (arrowheads). x 33,000. 262 K. TOSHIMORI, R. HIGASHI, AND C. 6 U R A +o m m ences were seen in junctional morphology except for the number of strands of type I1 zonula occludens on secretory cells: the control oviduct versus the ampulla (P < .05). Type I zonulae occludentes always had more strands than those of type 11. Further, type I zonulae occludentes had significantly more strands on secretory cells than on ciliated cells in each region-control oviduct, P < .01; isthmus, P < .001; ampulla, P < ,005. Type I1 zonulae occludentes had significantly more strands in secretory cells than in ciliated cells only in the control oviduct (P < .05). The distance between adjacent strands (mainly 30-50 nm in the adluminal regions) was similar for all regions at both diestrous and estrous stages. The depth occupied by junctional strands was a t least 0.51 & 0.20 pm in all zonulae occludentes. The depth was likely to be deeper at diestrous stage than at estrous stage. However, the minor differences in depth shown in Table 2 were not statistically significant. Maculae adherentes (desmosomes), which were displayed as characteristic patches of various-sized particles (Figs. 2, 3), were frequently observed below the occluding junctional domain in the epithelium at all regions of the oviduct. Nexuses (gap junctions) were rarely encountered in connection with the strands of zonula occludens in this epithelium (Fig. 1). However, many rectangularshaped clusters of small intramembraneous particles (diameter, about 60 A ) were sometimes observed below the occluding junctional domain or basalward in the epithelium (Fig. 2). NN 1 N I" N! 3 0 0 0 00 +I +I +I +I +I +I a m Ycq Dom 0 0 0 3 0 0 0 0 0 ncq " u?? ha, 0 0 N. m. v 3 0 0 0 t-m 11 +I +I +I +I +I +I 0 0 m 3 m a m N ?-! ,-!? ? 0 4 3 I i I 1 0 0 0 I I I 1 mm mm m o r?m N N c?? 30 00 00 m am i m I I I I worn m m i "t- *I *I 2 -2 .- 2; -dLd 23 8 C 0 8 4% $2 58 22 drd -ks 3 -3 aa W G ) &3 d N Lanthanum Infusion The lanthanum nitrate infused into the bloodstream permeated into the intercellular spaces between the epithelial cells in all regions. The tracer usually permeated the occluding junctional domain to within a very short distance from the luminal surface, and it could reach toward the luminal surface beyond the junctional domain in some regions (Fig. 6). DISCUSSION The present study demonstrates that the zonulae occludentes between oviductal epithelial cells are different in their geometrical structure between diestrous and estrous stages in the mouse: The anastomosing type is predominant a t the diestrous stage, and the parallel type is predominant at the es- ZONULAE OCCLUDENTES IN MOUSE OVIDUCT 263 Fig. 6 . Lanthanum infusion. The exogenous tracer permeates into the intercellular spaces. The tracer appears to he stopped at the juxtaluminal region (arrows), but it reaches toward the luminal surface in some regions (double arrows). X 7,400. Inset) High magnifica- tion of another juxtaluminal region. The tracer reaches toward the luminal surface beyond the occluding junctional domain (arrowhead). Some junctional elements (arrows) are outlined by tracer in negative contrast. x 49,000. trous stage. Since the reciprocal relationship between estradiol and progesterone levels in serum have been clearly elucidated during the 4-day estrous cycle in the rat (Kalra and Kalra, 19741, the progesterone dominant period is thought t o be coincident with the diestrous stage in this study, while the estrogendominant period is coincident with the estrous stage. Such a cyclic variation in morphology of zonulae occludentes is similarly reported in the rat uterine endometrium by Murphy et al. (1981). They stated that the frequently interlocked tight junctions (with many “t” intersections) were predominant when progesterone was administered to ovariectomized rats, while less frequently interlocked ones (with few “t” intersections) were predominant when estrogen was adminstered. Since hormonal change affects not only uterine endometrial epithelium but also oviductal epithelium, it seems reasonable that the present result is in accord with that of Murphy et al. (1981). However, these data do not correspond in all respects to those reported by Komatsu et al. (1979),who stated that the zonulae occludentes between secretory cells were anastomosing, while those between ciliated cells were parallel in mouse oviduct. In our opinion, the structural type of zonula occludens fluctuates according to the ovarian hormone level in serum. Therefore, we agree with the concept that tight junctional configurations are dynamically interchangeable (or reconstructable) under various conditions (Pitelka et al., 1973; Humbert et al., 1976; Montesano et al., 1976; Suzuki and Nagano, 1978). The present data showing that the depth occupied by junctional domain at the diestrous stage is likely to be deeper than that at 264 K. TOSHIMORI, R. HIGASHI, AND C. 6URA the estrous stage make us agree with Murphy et al. (1981), who reported that the tight junctions extended more deeply down in the rat uterine endometrium when progesterone was administered. The zonulae occludentes between oviductal epithelial cells are thought to be tight enough morphologically to seal intercellular spaces, since they belong to the category of tight type according to a classification proposed by Claude and Goodenough (1973). The sealing effect of such zonulae occludentes is supposed not to be seriously affected by ovarian hormones at any region in the mouse oviduct throughout the estrous cycle. On the other hand, the result of lanthanum infusion suggests that the zonulae occludentes in the oviductal epithelium do not always function as a diffusion barrier to the exogenous tracer. Martinez-Palomo and Erlij (1975) reported that the permeability of zonulae occludentes was not directly correlated with the number of junctional strands. The labeled proteins identical or similar to those in serum were confirmed to be transferred from the ampulla of the oviduct to the developing embryo, while reverse transfer from the lumen to the oviductal epithelium did not occur in mice (Glass, 1969). Such serum type proteins are supposed to be due to a combination of a transudate from bloodstream and an active secretion of secretory cells (Mastroianni et al., 1970; Moghissi, 1970). Further, immunoglobulins have been confirmed to be present in the human oviduct, and they are postulated to have a role in immunologic infertility (Moghissi, 1970). Lanthanum nitrate is currently considered to be a “small” tracer in which it is impossible to evaluate the actual size and molecular weight due to a charged molecule binding to various substances in the intercellular space (Tice et al., 1977). Therefore, an experimental probe using such exogeneous tracers as horseradish peroxidase (moJecular weight, 40,000; diameter, about 50 A), microperoxidase (molecular weight, 1,900; diameter, about 20 A), and 5-hydroxydopamine (molecular weight, 256, diameter, 5-7 A ) is necessary to assess the permeable molecular weight through intercellular spaces a t various regions of the oviduct during the estrous cycle. Further, the exogeneous tracers need to be applied under physiological conditions. Exogeneous tracers including lanthanum nitrate can alter its permeability among differ- ent epithelia under various experimental conditions (Brightman and Reese, 1969; Goodenough and Revel, 1970; Friend and Gilula, 1972; Machen et al., 1972; Wade et al., 1973; Martinez-Palomo and Erlij, 1975). Since lanthanum nitrate was applied along with fixative in this study, its increased degree of penetration is not a reliable reflection of permeability in the physiological state. The zonulae occludentes are thought not only to store the luminal fluid that provides a medium for the passage of spermatozoa and ovulated ova through the oviduct but also to prevent the escape of spermatozoa and ovulated ova or their degradation products from the lumen to the extracellular space (or to the bloodstream), which would induce a n immune response (production of antibodies to spermatozoa and autoantibodies to ova). One could speculate from the number of parallel- or anastomosing-type zonulae occludentes (Table 1)that secretory cells are more effective barriers than ciliated cells in the isthmus, while the reverse is true in the ampulla. It is of interest that this correlates with the fact that secretory cells are more numerous in the isthmus, while ciliated cells are more numerous in the ampulla (Clyman, 1966; Hafez, 1973). The patterns of occluding junctions are related to variations in the cell shape (Pitelka et al., 1973; Hull and Staehelin, 1976; Suzuki and Nagano, 1978). In the mouse oviduct, secretory cells increase in height as they increase in secretory activity during the estrous stage (Beier, 1974).Further, Suzuki and Tsutsumi (1981) reported in the normally mated rabbit that intraluminal pressure of the oviduct increased during estrous and then declined gradually during the period of egg transport through the isthmus. Therefore, when the idea proposed by Hull and Staehelin (1976) is adapted to the present results, the zonulae occludentes in the oviduct may become flexible during estrous stage due to increased stress from luminal contents and cytoplasm under the influences of estrogen, while they may become less flexible during diestrous stage due to decreased stress after resorption of luminal contents and cytoplasm under the influence of progesterone, It is also reasonable that adjacent epithelial cells are tightly conjoined by many maculae adherentes to endure the vigorous movement of cilia throughout estrous cycle, since maculae adherentes, which are well ZONULAE OCCLUDENTES IN MOUSE OVIDUCT known to serve as cell-to-cell adhesion (Farquhar and Palade, 1963; Friend and Gilula, 1972; Staehelin, 19741, are frequently observed in the oviductal epithelium. Further, the occluding junctions are also implicated in the adhesive function as well as providing a permeability seal (Farquhar and Palade, 1963; Brightman and Reese, 1969; Pitelka et al., 1973; Montesano et al., 1975; Pinto da Silva and Kachar, 1982). However, it is still unclear whether or not adjacent epithelial cells are well coupled ionically or metabolically throughout the estrous cycle, since nexuses, which are well known as communicating junctions (Peracchia, 1980; Hertzberg et al., 1981), are sometimes, but not frequently, discernible in the oviductal epithelium. ACKNOWLEDGMENTS The authors thank Dr. R.J. Adams for his advice on English usage. This work was supported by Grants-Aid for Scientific Research from the Ministry of Education, Science, and Culture of Japan (No. 557009, No. 56770019, No. 57770026). LITERATURE CITED Bareither, M.L., and H.G. Verhage (1981) Control of the secretory cell in cat oviduct by estradiol and progesterone. Am. J. Anat., I62:107-118. Beier, H.M. (1974) Oviducal and uterine fluids. J. Reprod. Fertil., 37,221-237. Blandau, R.J., ed. (1971) The Biology of the Blastocyst. University of Chicago Press, Chicago and London. Brightman, M.W., and T.S. Reese (1969) Junctions between intimately apposed cell membranes in the vertebrate brain. J. Cell Biol., 40:648-677. Claude, P., and D.A. Goodenough (1973) Fracture faces of zonulae occludentes from “tight” and “leaky” epithelia. J. Cell Biol., 58:390-400. Clyman, M.J. (1966) Electron microscopy of the human fallopian tube. Fertil. Steril., 17,281-301. Farquhar, M.G., and G.E. Palade (1963) Junctional complexes in various epithelia. J. Cell Biol., 17:375-412. Friend, D.S., and N.B. Gilula (1972) Variations in tight and gap junctions in mammalian tissues. J. Cell Biol., 53758-776. Glass, L.E. (1969) Immunocytological studies of the mouse oviduct. In: The Mammalian Oviduct. E.S.E. Hafez and R.J. Blandau, eds. University of Chicago Press, Chicago, pp. 459-476. Goodenough, D.A., and J.P. Revel (1970) A fine structural analysis of intercellular junctions in the mouse liver. J. Cell Biol., 45t272-290. Hafez, E.S.E. (1973) Transport of spermatozoa in the female reproductive tract. Am. J. Obstet. Gynecol., 115,703-717. Hertzberg, E.L., T.S. Lawrence, and N.B. Gilula (1981) Gap junctional communication. Annu. Rev. Physiol., 43t479-491. Hull, B.E., and L.A. Staehelin (1976) Functional significance of the variations in the geometrical organization 265 of tight junction networks. J. Cell Biol., 68,688-704. Humbert, F., R. Montesano, A. Perrelet, and L. Orci (1976) Junctions in developing human and r a t kidney: A freeze-fracture study. J. Ultrastruct. Res., 56:202214. Inoue, S., and J.C. Hogg (1977) Freeze-etch study o f the tracheal epithelium of normal guinea pigs with particular reference to intercellular junctions. J. Ultrastruct. Res., 6139-99. Kalra, S.P., and P.S. Kalra (1974) Temporal interrelationships among circulating levels of estradiol, progesterone and LH during the rat estrous cycle: Effects of exogenous progesterone. Endocrinology, 95,1711-1718. Komatsu, M., and H. Fujita (1978) Electron microscopic studies on the development and ageing of the oviduct epithelium of mice. Anat. Embryol., 152,243-259. Komatsu, M., K. Ishimura, and H. Fujita (1979) Freezefracture images of the zonula occludens in the mouse oviduct epithelium. Arch. Histol. Jpn., 41:453-458. Machen, T.E., D. Erlij, and F.B.P. Wooding (1972) Permeable junctional complexes. The movement of lanthanum across rabbit gallbladder and intestine. J. Cell Biol., 54:302-312. Martinez-Palomo, A,, and D. Erlij (1975) Structure of tight junctions in epithelia with different permeability. Proc. Natl. Acad. Sci. (U.S.A.),72:4487-4491. Mastroianni, L., Jr., M. Urzua, and R. Stambaugh (1970) Protein patterns in monkey oviductal fluid before and after ovulation. Fertil. Steril., 21:817-820. Moghissi, K.S. (1970) Human fallopian tube fluid. I. Protein composition. Fertil. Steril., 212321-829. Montesano, R., D.S. Friend, A. Perrelet, and L. Orci (1975) In vivo assembly of tight junctions in fetal r a t liver. J. Cell Biol., 67t310-319. Montesano, R., G. Gabbiani, A. Perrelet, and L. Orci (1976) In L I ~ U Oinduction of tight junction proliferation in r a t liver. J. Cell Biol., 68t793-798. Murphy, C.R., J.G. Swift, T.M. Mukherjee, and A.W. Rogers (1981) Effects of ovarian hormones on cell membranes in the r a t uterus. 11. Freeze-fracture studies on tight junctions of the lateral plasma membrane of the luminal epithelium. Cell Biophys., 3:57-69. Odor, D.L., P. Godduni-Rosse, R.E. Rumery, and R.J. Blandau (1980) Cyclic variations in the oviductal ciliated cells during the menstrual cycle and after estrogen treatment in the pig-tailed monkey, Macaca nemestrina. Anat. Rec., 198:35-57. Peracchia, C. (1980)Structural correlates of gap junction permeation. Int. Rev. Cytol., 66531-146. Pinto da Silva, P., and B. Kachar (1982) On tight-junction structure. Cell, 28:441-450. Pitelka, D.R., S.T. Hamamoto, J.G. Duafala, and M.K. Nemanic (1973) Cell contacts in the mouse mammary gland. I. Normal gland in postnatal development and the secretory cycle. J. Cell Biol., 56;797-818. Reese, T.S., and M.J. Karnovsky (1967) Fine structural localization of a blood-brain barrier with exgeonous peroxidase. J. Cell Biol., 34.207-217. Staehelin, L.A. (1974) The structure and function of intercellular junctions. Int. Rev. Cytol., 39t191-283. Suzuki, F., and T. Nagano (1978) Development of tight junctions in the caput epididymal epithelium of t h e mouse. Dev. Biol., 63t321-334. Suzuki, H., and Y. Tsutsumi (1981) Intraluminal pressure changes in the oviduct, uterus, and cervix of the mated rabbit. Biol. Reprod., 24t723-733. Tice, L.W., R.C. Carter, and M.C. Cahill (1977) Tracer and freeze fracture observations on developing tight junctions in fetal r a t thyroid. Tissue Cell, 9:395-417. Tice, L.W., S.H. Wollman, and R.C. Carter (1975) 266 K. TOSHIMOHT, R. HIGASHI, AND C. OUKA Changes in tight junctions of thyroid epithelium with changes in thyroid activity. J. Cell Biol., 66:657-663. Toshimori, K. 11982) Penetration of the mouse sperm head through the zona pellucida in vivo: An electromicroscope study a t 2_00KV.Biol. Reprod., 26:475-481. Toshimori, K., and C. Oura (1982) Cellular interconnections in the young mouse ovary. Freeze-fracture study. Cell Tissue Rcs., 224:383-395. Toshimori, I<., R. Higashi, and C. Oura (1982). In viuo fertilization of mouse. V. Observations on the zonulae occludentes bet ween oviductal epithelial cells-frcezefracture study. J. Electron Microsc., 31:331 (Abstract). Urzua, M.A., R. Stambaugh, G. Flickinger, and L. Mastroianni, J r . (19701 Uterine and oviduct fluid protein patterns in the rabbit before and after ovulation. Fertil. Steril., 21:860-865. Verhage, H.G., M.L. Bareither, R.C. Jaffe, a n d M . Akbar (1979) Cyclic changes i n ciliation, secretion and cell height of t h e oviductal epithelium in women. Am. J. Anat., 156r505-522. Wade, J.B., J. P. Revel, and V.A. DiScala (19731Effect of osmotic gradients on intercellular junctions of the toad bladder. Am. J. Physiol., 224:407-415. West, N.B., H.G. Verhage, and R.M. Brenner (1976) Suppression of the estradiol receptor system by progesterone in the oviduct and uterus of the cat. Endocrinology, 99: 1010-1016.