Rat uterine tissue and cell responses to the presence of plain and indomenthacin-delivering IUDs.код для вставкиСкачать
THE ANATOMICAL RECORD 208507-514 (1984) Rat Uterine Tissue and Cell Responses to the Presence of Plain and Indomethacin-DeliveringIUDs C.C. PAGE, P.R. HURST, AND G.F.S. SPEARS Departments of Anatomy, (C. C.E?, I!R. H.) and Social and Preuentiue Medicine (G.RS.S.), Faculty of Medicine, University of Otago, Dunedin New Zealand ABSTRACT Plain silastic intrauterine devices or those containing 270 pg of indomethacin were inserted into the caudal portion of one uterine horn of mature Wistar rats. After a 3-week period animals were fixed by perfusion on the morning of day 2 after estrus. Segments of uterine tissue corresponding to regions adjacent to and cranial to the devices as well as an equivalent portion of the contralateral horn were embedded in glycol methacrylate. A group of control animals without any form of device were treated in an identical manner. Sections cut from these segments were evaluated by grid-point stereology to ascertain changes in tissue volumes and cell populations. It was found that the presence of plain devices induced hypertrophy in the stroma and myometrium of the portion of the uterus adjacent to the device. The presence of indomethacin in such devices prevented stromal hypertrophy. No changes in populations of fibroblasts or areas of glandular or vascular tissue were evident in any treatment group. Cell populations of neutrophils, eosinophils, and mononuclear cells, however, were elevated in the superficial stroma of the horns bearing either type of device; this feature was more pronounced for neutrophils in the presence of the indomethacin devices. Neutrophils, rather than eosinophils, predominated in the epithelia of the uterus bearing either type of IUD. Conversely, eosinophil populations were reduced in the superficial tissues cranial to the devices delivering indomethacin. Neutrophils and mononuclear cells were also found to be elevated in the deep stroma of tissues adjacent to both the plain and medicated device. The accumulation of inflammatory cells in the endometrium in response to the presence of intrauterine devices (IUDs) as well as their emigration into the uterine lumen has been commonly observed in experimental animals such as the rabbit (El Sahwi and Moyer, 19711, rat (Doyle and Margolis, 19631, and rhesus monkey (Marston et al., 1971; Hurst et al., 1977). This response is one of many seen in the human uterus after insertion of IUDs (Sheppard and Bonnar, 1980). The studies involving observations on uterine tissue were, however, qualitative and, apart from attempts at quantitations of changes to luminal cells (El Sahwi and Moyer, 1971) or human endometrial vascularity (Shaw et al., 1979), quantitative studies are lacking on changes that might occur in total uterine tissue as well as cell distributions in uteri bearing IUDs. The present study has therefore attempted to quantify total tissue changes and patterns 0 1984 ALAN R. LISS, INC. of cellular distribution within the rat uterus and ascertain what changes occur after fitting rats with IUDs for 3 or 4 weeks. Two types of device were selected, a rod-shaped silastic device covered by a vinyl sleeve, and a similar system loaded with indomethacin which is designed to deliver low levels (4-5 pglday) of this anti-inflammatory drug (Peplow and Hurst, 1981). It is known that in cyclic or late pregnant rats the population of stromal macrophages, eosinophils, and neutrophils fluctuates considerably (Rytomaa, 1960; Ross and Klebanoff, 1966; Padykula, 1981).Thus uterine tissue was studied in control and IUD-fitted rats which were all selected on the morning of day 2 of metestrus. Received June 10, 1983; accepted September 29, 1983. Address reprint requests to Dr. P.R. Hurst, Anatomy Department, University of Otago, Dunedin, New Zealand. 508 C.C. PAGE, P.R. HURST, AND G.F.S. SPEARS MATERIALS AND METHODS Three groups of five or six Wistar rats, age 4 months, were used in this study. Group 1 animals served as controls and carried no IUDs. Group 2 animals were fitted with plain silastic IUDs, and group 3 were fitted with IUDs containing 270 pg indomethacin a t laparotomy under pentobarbitone anaesthesia on day 2 of metestrus (Peplow and Hurst, 1981). The IUDs measured 1.0 X 0.1 cm and were inserted into the lower half of the left uterine horn. The animals were then rested for 23-29 days during which time daily vaginal smears were taken. Only those rats showing consistent estrous cycles of 4 days’ duration were retained in the study. On day 2 of metestrus following this rest period the animals were anaesthetised and the uterine tissue fixed in situ by aortic perfusion with saline (for about 1rnin), followed by 10% phosphate-buffered formalin a t a pressure of 100 mm Hg. The uterine horns were divided into three regions (Fig. 1): A) immediately rostra1 to the IUD; B) adjacent to the IUD, the device being retained during preparation; and C) from the contralateral horn in a position corresponding to B. In group 1 (animals without a n IUD) the tissue was divided in a corresponding manner. Tissue from each of these regions, except where B contained a n IUD (groups 2 and 31, was divided into three segments before embedding. The tissue was kept in 10% formalin until it was required. To prepare sections for light microscopy each piece of tissue was dehydrated through 70%, 95%, and absolute alcohol followed by Fig. 1. Female rat reproductive system with silastic IUD. three changes of glycol methacrylate infiltrating solution before being embedded under vacuum in glycol methacrylate. Tissue containing a n IUD required longer dehydrating and infiltrating periods. After embedding, tissue containing a n IUD was also cut into three segments and reembedded. Therefore, for each animal there were nine blocks of tissue: Al, A2, As, B1, B2, B3, C1, C Z , and Cs. Sections (2 pm) were cut on a Sorvall JB/4 microtome and six slides, each containing three consecutive sections, were prepared from each of the nine segments. The six sets of sections were made a t about 50 pm intervals along each region of the uterine horns. Sections were stained for 20 sec in filtered Lillie Mayer hematoxylin, washed for 5 min in running water, and then stained for 60 min in fresh, filtered Wrights Blood stain buffered to pH 4.8 with 0.1 M acetate buffer. This particular staining procedure allows clear distinction of neutrophils and eosinophils, with the latter being strongly eosinophilic at a pH of less than 5. The slides were then dehydrated in three changes of absolute alcohol (1-2 min each change), cleared in xylene, and mounted. Slides were coded so that the observer did not know the origin of the sections. Grid-point stereology (Elias and Hyde, 19801, using a 100-point grid overlay on a Reichert visopan projection microscope, was employed to determine volume fractions of stromal fibroblasts, eosinophils, neutrophils, blood vessels including the lumen and wall tissue of capillaries, glands, and leucocytes in the epithelium, as well as total proportions of muscle, epithelium, stroma, lumen, and total tissue. Eight random fields of endometrium were evaluated on each section, four in a zone including a small part of the epithelium (superficial stromal zone) and four in a zone close to the muscle layer (deep stromal zone). Using a lower power of magnification ( x 251, proportions of muscle, stroma, epithelium, and lumen were determined; adventitia was disregarded. Volume fractions were determined by the formula Vv = Pn/Pt, where Pn = number of grid points overlying a specific tissue or cell, and Pt = the total number of points on the grid. This information was entered into a B5930 computer and analysed using a n analysis of variance programme from the Statistical Package for the Social Sciences (Nie et al., 1975)followed by a Scheffe analysis (Scheffe, UTERINE RESPONSES TO IUDs IN RATS 1959)of group differences. This involved comparing the mean volume fraction for each variable for each region (A, B, or C) of uterus with other regions within the same groups of animals and with the same region in the other two groups of animals. Differences in means were taken to be statistically significant a t a level of 1%or less. RESULTS The most significant findings are shown in Figures 2 and 10 and summarised in the corrresponding tables. For all groups the volume of tissue (Fig. 2) in region A was less than the total tissue volume of region B and region C because the uterine horns taper cranially (Fig. 1).The volume of region C remained constant among the three groups. There was a significant increase in total tissue volume in region B of groups 2 and 3 (Table 1).In group 2 this increase was due to elevated volumes of both stroma and muscle whereas in group 3 it was due to a n increased amount of muscle only; the proportion of stroma remained similar to that of region B in animals not fitted with IUDs (Figs. 3, 4; Table 2). Volumes of epithelium were also estimated, but these showed no significant changes owing to the various treatments. The presence of either type of IUD caused the lumen to be distended. Fibroblasts were easily distinguished by their fine cytoplasmic processes, large pale nuclei, and prominent, dark-stainingnucleoli (Fig. lla); however, no changes were found in the volume fractions of these cells. Also, no alterations were evident in the volume fractions of blood vessels and glands in the stroma of uteri bearing either plain or indomethacin-loaded IUDs. In group 1 (no IUD), there were no differences in the distribution of polymorphonuclear or mononuclear cells in any region or zone. TABLE 1. Statistically significant differences at the 1% level by a one-way analysis of variance followed by Scheffk's test Between regions within groups: Stroma and muscle B > A = C Group 2: Total tissue B > A = C Group 3: Muscle B > A = C Total tissue B > A = C (Stroma B > A = C a t P = 0.02) Between groups within regions: Region A: Stroma 2 > 1 = 3 Muscle 2 > 3 > 1 Muscle 2 > 3 > 1 Region B: Total tissue 2 > 1 fig. 2 lZo rTot.1 T T " 0 g 45 509 ABC ABC ABC no plain ind IUD IUD IUD 40 35 30 25 n " ABC ABC ABC no plain ind IUD IUD IUD Figs. 2-4. Low-power magnification estimations (+ SEMI of total tissue (21, relative stromal volume (31, and relative muscle volume (4)of uterine tissues of control and IUD groups for the different regions A, B, and C as depicted in Figure 1. 5 10 C.C. PAGE, P.R. HURST, AND G.F.S. SPEARS TABLE 2. Mean proportion of stroma and muscle in total tissue volume (within groups)' Region Percent stroma Percent muscle A No IUD (group 1) B C Plain IUD (group 2) A B C I A Ind. IUD (group 3) B C 60.0 60.6 58.1 60.0 56.9 59.1 58.2 54.1 60.8 39.0 38.5 40.5 37.8 40.7 39.0 40.0 43.7 37.4 'Data derived from Figures 2-4 TABLE 3. Significant differences in proportional cell volumes between regions A, B, and C within groups I , 2, and 3 Superficial endometrial zone: Group 1 No significant differences Epithelial leucocytes Group 2 Stromal neutrophils Stromal eosinophils Group 3 Epithelial leucocytes Stromal neutrophils Mononuclear cells Deep endometrial zone: All groups No significant differences B>C=A A>B>C A=B>C B>A>C B>A=C A=B>C In animals with IUDs, changes were observed in the leucocyte populations (Table 3) as follows: Epithelial Leucocytes (Figs. 5, 11) More than 99% of these were neutrophils, only a few eosinophils being sighted in the epithelium of the hundreds of sections evaluated. There were increases in the occurrence of epithelial leucocytes in region B of both groups carrying IUDs, and also in region A of the group with the indomethacinloaded IUD. In the contralateral uterine horns (region C), there were no significant differences in the proportion of epithelial leucocytes between any of the three groups. In numerous sections of regions A and B leucocytes were observed in the lumen; these appeared to be predominately degenerated neutrophils, but were not subject to quantitation. Eosinophils were also evident in the lumen. TABLE 4. Significant differences in proportional cell volumes between groups 1, 2, and 3 within Pegions A, B, and C Superficial endometrial zone: Region A Epithelial leucocytes Stromal neutrophils Stromal eosinophils Mononuclear cells Region B Epithelial leucocytes Stromal neutrophils Stromal eosinophils Mononuclear cells Region C Stromal neutrophils Mononuclear cells Deep endometrial zone: Stromal neutrophils Region A Mononuclear cells Region B Stromal neutrophils Mononuclear cells Reeion C Mononuclear cells 3 2 2 3 > = > > 1 3 1 2 = 2 > 1 = 3 > 1 2 3 2 3 = > = > 3 2 3 1 > > > = 1 1 1 2 3 > 1 = 2 3 > 1 = 2 2 2 2 3 3 = = = > > 3 3 3 1 1 > > > = = 1 1 1 2 2 in the region adjacent to the IUD only (Table 4; Fig. 6). No difference was seen between any region of superficial stromal zone eosinophils of control animals (group 1)or the contralateral horns of animals bearing either type of IUD. In the deeper zone of stroma adjacent to the muscle there were no significant differences in eosinophil populations between regions of uterine tissue, or between any groups of animals. Neutrophils Neutrophil populations were elevated in the superficial stroma of the uterine horns bearing either plain or indomethacin-loaded Stromal Leucocytes IUDs (Fig. 7). This increase was more marked Eosinophils in the areas directly adjacent to the IUDs In the superficial zone of stroma a marked and in the region above the plain IUD (Table increase in eosinophils was observed in both 4). In the deeper stroma neutrophils were regions A and B of the uterine horns bearing again elevated by the presence of either type plain IUDs (group 2). For the uteri bearing of IUD (Fig. 81, although their density was an indomethacin-loadedIUD (group 31, eosin- less when compared with an equivalent volophil populations were significantly elevated ume of tissue in the superficial stroma. 511 UTERINE RESPONSES TO IUDs IN RATS fig. 5 fig. 6 T - Stromal ioninuphila - 0.2 -s c fig.? Stromal 1.2 fig.9 ,-- fig.10 Stromal Mononuclear 0.8 0.4 0 no plain IUD IUD ind IUD Figs. 5-10. High-power magnification estimations of the mean volume fractions of various cell types in uterine tissues of the three groups. Fig. 5) Leucocytes in the epithelium; 6) eosinophils in the superficial stroma; 7) Mononuclear Cells (Figs. 9-11) It was noted in both superficial and deep stroma that there were isolated mononuclear cells which were likely to be either lymphocytes, plasma cells, or macrophages. The staining procedure adopted here was primarily used to demonstrate either neutrophils or eosinophils, but unfortunately these mononuclear cells were not always specifically identifiable as lymphocytes, macrophages, or plasma cells. It was found that these cells were elevated in horns bearing either type of IUD, although this increase was more marked by the presence of a n indomethacinloaded IUD in both zones of stroma studied. There was a threefold increase of these cells ABC A B C ABC no plain ind I U D IUD IUD neutrophils in the superficial stroma; 8) neutrophils in the deep stroma; 9) mononuclear cells in the superficial stroma; 10)mononuclear cells in the deep stroma. in the contralateral horn of the rats fitted with a n indomethacin-loaded IUD when compared with the control group. DISCUSSION It has recently been reported that nonmedicated IUDs, made in a similar manner to those used in the present study, caused a n increase in uterine tissue weight and that the incorporation of indomethacin diminishes this hypertrophy (Peplow-and Hurst, 1982). Evaluation of tissue areas, in the present study, confirms this finding, and furthermore has determined that both stroma and muscle are increased by a plain IUD in the rat. In a study of 14 cases, myometrial hypertrophy was also observed in 12 human IUD- 512 C.C. PAGE, P.R. HURST, AND G.F.S. SPEARS Fig. 11. Superficial endometrium (a)from control animal and (b) adjacent to an indomethacin-loadedIUD. In (b) a portion of the IUD is seen separated from the epithelium by a small area of lumen (L).Small arrows indicate epithelial neutrophils; large arrows show neutrophils at the stromal/epithelial boundary. F = fibroblasts; M = mononuclear cells (see text). Magnification, x860. UTERINE RESPONSES TO IUDs IN RATS bearing uteri (Honore, 1979). The finding here that only the muscle component is elevated when such IUDs deliver 4-5 pg/drug/ day (Peplow and Hurst, 1981) suggests that when delivered from a n intraluminal site, low levels of the drug are capable of suppressing stromal hypertrophy. It remains to be determined if higher doses can influence the muscle layers that lie deeper to the source of indomethacin. The elevation of neutrophil numbers within the epithelium of uterine horns fitted with either type of IUD suggests a unidirectional migration of these cells toward the IUD. This finding supports many previous observations on IUD action (reviewed in Duncan and Wheeler, 1975; Moyer et al., 1976). It was apparent that the increase in epithelial leucocytes was most predominant in the region of tissue adjacent to the IUDs. Nonsteroidal anti-inflammatory drugs are reportedly inhibitory to leucocyte migration (see below), and yet higher numbers were seen in the epithelium cranial to the indomethacin device than in the same area above the plain IUD. Most of the epithelial leucocytes were neutrophils, suggesting either that eosinophils do not predominate in the foreign body reaction or their migration rate was submaximal at the time of sampling after the insertion of the devices. Similarly, Padykula and Tansey (1979) showed that heterophils (neutrophils) and monocytes, rather than eosinophils, were elevated in the superficial endometrium during late pregnancy in the rat. It should be borne in mind, however, that degranulation of eosinophils may have occurred during their passage through the basal lamina and epithelium; if this had happened, the staining procedure used here would have equivocally identified them only in the stroma. It is known that eosinophils can modulate inflammation by exocytosis of their granule contents into extracellular spaces (Beeson and Bass, 1977) and undergo lysis in the stroma of rats during estrus (Ross and Klebanoff, 1966). Further studies are clearly needed, perhaps by a n ultrastructural analysis, in order to clarify this as well as a determination of changes in the epithelial and luminal leucocyte population at various times after IUD insertion. Concomitant with the increase in epithelial leucocytes, a marked rise in stromal neutrophils was a distinct feature of the tissue response close to both forms of IUD. Almost twice the volume of these cells was seen in 513 the uterine horns bearing the indomethacin device. This increased accumulation might represent a greater response by neutrophils to indomethacin, in which case the drug would not be preventing this part of a n inflammatory reaction. Alternatively, a slower migration was occurring once these cells had left the vascular system, resulting in a buildup of larger numbers to be observed at the time of sampling. This latter possibility would be favoured if tissue neutrophil chemotaxis is inhibited by nonsteroidal antiinflammatory drugs in the uterus as is reported for leucocytes in other tissues (Di Rosa, 1979) or from in vitro studies (Spisani et al., 1979). Contrary to the finding of extensive numbers of neutrophils throughout the stroma in response to the different IUDs, eosinophils were increased only in the superficial zone of the stroma, with indomethacin reducing this feature cranial to the device. The similarity of eosinophil volumes in the stroma close to the muscle zone of all groups suggests again that IUDs do not influence this cell type in the deep endometrium, and that those that were seen in increased frequency in the superficial stroma had emigrated from proximate vascular tissue. Prostaglandins are important mediators of increased vascular permeability, and it is well established that indomethacin is a potent inhibitor of cyclo-oxygenase causing a suppression of prostaglandin production (Vane, 1971; Gryglewski, 1979). If IUDs are causing increased vascular permeability (Shaw et al., 19791, it might be expected that a n indomethacin delivery system would reduce this and consequently edema would not be so evident. If there had been edema there would probably have been a reduction in the volume fraction of fibroblasts, since the extracellular compartment would have been increased. This study showed a constancy of fibroblast populations with respect to tissue volume in all treatment groups, suggesting that edema is not a major consequence of the insertion of IUDs in rats. Monocytes, macrophages, and plasma cells have been shown to be elevated in the stroma of postpartum rats (Padykula and Campbell, 1976), and monocytes were found in the epithelium and to accumulate at the stromalepithelial boundary during late pregnancy (Padykula and Tansey, 1979). Macrophages were clearly identified in monkey uterine luminal flushes from IUD-fitted animals (Hurst 514 C.C. PAGE, P.R. HURST, AND G.F.S. SPEARS et al., 1977) and formed a substantial population on human IUDs (Myatt et al., 1975). The study reported here tentatively suggests that some or all of these cell types might also be undergoing changes in the stroma, but it must be emphasised that a further study is required to ascertain more precisely if any alteration t o the population of these cells is occurring in response to the presence of plain and medicated IUDs. ACKNOWLEDGMENTS This work was performed whilst one of us (C.C.P.)held a summer research scholarship from the Otago Medical Research Foundation. We are grateful to Vicki Gamble, Keith Pickersgill, and Ken Turner for their technical assistance, and Margaret Scoular for typing the manuscript. LITERATURE CITED Beeson, P.B., and D.A. Bass (1977)The Eosinophil. W.B. Saunders, Philadelphia. Di Rosa, M. (1979) Inhibition of cell migration in vivo and granuloma formation. In: Anti-inflammatory Drugs. J.R. Vane and S.H. Ferreira, eds. SpringerVerlag, New York, pp. 223-254. Doyle, L.L., and A.J. Margolis (1963) Intrauterine foreign body-effect on pregnancy in the rat. Science, 139:833-834. Duncan, G.W., and R.G. Wheeler (1975)Pharmacological and mechanical control of implantation. Biol. Reprod., 12143-175. El Sabwi, S., and D.L. Moyer (1971)The leukocytic response to a n intrauterine foreign body in the rabbit. Fetil. Steril., 22398-408. Elias, H., and D.M. Hyde (1980)An elementary introduction to stereology (quantitative microscopy). Am. J. Anat., 159t412-446. Gryglewski, R.J. (1979)Screening and assessment of the potency of anti-inflammatory drugs in vitro. In: Antiinflammatory Drugs. J.R. Vane and S.H. Ferreira, eds. Springer-Verlag,New York, pp. 3-43. HonorB, L.H. (1979) Menorrhagia, diffuse myometrial hypertrophy and the intrauterine contraceptive device: A report of fourteen cases. Acta Obstet. Gynecol. Scand., 58283-285. Hurst, P.R., K. Jefferies, P. Eckstein, K. Dawson, and A.G. Wheeler (1977)Leukocytes are consistently associated with degenerating embryos in IUD-bearing rhe. sus monkeys. Nature, 269r331-333. Marston, J.H., W.A. Kelly, and P. Eckstein (1971) Deciduomal reaction induced in rhesus monkeys by tuba1 injection of arachis oil and by the presence of a n intrauterine device. J. Reprod. Fertil., 25:451-454. Moyer, D.L., S.T. Shaw, and J.C. Fu (1976)Mechanisms of action of intrauterine contraceptive devices. In: Regulation of Human Fertility. K.S. Moghissi and T.N. Evans, eds. Wayne State University Press, Detroit, pp. 174-193. Myatt, L., M.A. Bray, D. Gordon, and J. Morley (1975) Macrophages on intrauterine contraceptive devices produce prostaglandins. Nature, 257:227-228. Nie, N.H., C.H. Hull, J.G. Jenkins, K. Steinbrennek, and D.H. Bent (1975) Statistical Package for the Social Sciences. McGraw-Hill, New York. Padykula, H.A. (1981)Shifts in uterine stromal cell populations during pregnancy and regression. In: Cellular and Molecular Aspects of Implantation. S.R. Glasser and D.W. Bullock, eds. Plenum, New York, pp. 197216. Padykula, H.A., and A. Campbell (1976) Cellular mechanisms involved in cyclic stromal renewal of the uterus. 11. The albino rat. Anat. Rec., 18427-48. Padykula, H.A., and T.R. Tansey (1979)The occurrence of uterine stromal and intraepithelial monocytes and heterophils during normal late pregnancy in the rat. Anat. Rec., 193t329-356. Peplow, P.V., and P.R. Hurst (1981)An intrauterine system for the sustained release of indomethacin. Prostaglandins Med., 7563-569. Peplow, P.V., and P.R. Hurst (1982) Influence of local release of indomethacin from a silastic device on uterine prostaglandins F2" and E. Prostaglandins Leukotrienes Med. 9:437-444. Ross, R., and S.J. Klebanoff (1966)The eosinophil leukocyte. Fine structure studies on changes in the uterus during the estrous cycle. J. Exp. Med., 124,653-660. Rytomaa, T (1960)Organ distribution and histochemical properties of eosinophil granulocytes in rat. Acta Patho1 Microbiol. Scand. Suppl. 140,50:1-72. Shaw, S.T., L.K. Macaulay, and W.R. Hohman (1979) Vessel density in endometrium of women with and without intrauterine contraceptive devices: A morphometric evaluation. Am. J. Obstet. Gynecol., 135.202-206. SheffB, H.A. (1959) The Analysis of Variance. Wiley, New York. Sheppard, B.L., and J. Bonnar (1980) The response of endometrial blood vessels to intrauterine contraceptive devices: An electron microscopic study. Br. J . Obstet. Gynaecol., 87:143-154. Spisani, S., G. Vanzini, and S.Traniello (1979)Inhibition of human leukocytes locomotion by anti-inflammatory drugs. Experientia, 35803-804. Vane, J.R. (1971) Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nature New Biol., 231-237.