LEUCOCYTOGENESIS AND BACTERICIDAL IN GENITAL TRACT TISSUES OF THE ESTROUS AND PSEUDOPREGNANT RABBIT TISSUE ACTIVITY Thesis for the Dogma of M. S. MICHIGAN STATE UNIVERSITY Charles R. Wira 196‘6 $315,313 1..“ ‘- L ‘ .44“- n 1—w~‘ LTRPADV M “11 164‘ L w University TISSUE LEUCOCYTOGENESIS AND BACTERICIDAL ACTIVITY IN GENITAL TRACT TISSUES OF THE ESTROUS AND PSEUDOPREGNANT RABBIT By Charles R. Wira A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Physiology 1966 "A mind, nimble and versatile enough to catch the resemblances of things, and at the same time steady enough to fix and discern their subtle differences." Francis Bacon (1650) ACKNOWLEDGMENTS I would like to take this opportunity to express my sincere gratitude to Dr. John E. Nellor for the valuable assistance, criticism, and widsom which he afforded me during this course of study. It is with heartfelt thanks that I express my appreciation to him for his guidance and advice as both teacher and friend. My special thanks are due to Dr. David A. Reinke and Dr. Raymond F. Johnston for their constructive criticism and cooperation in making this a more meaningful thesis. Appreciation is also extended to Dr. Bernard V. Alfredson, chairman of the Department of Physiology, Dr. Gail Riegle, and Dr. Clyde F. Cairy for their suggestions and advice. I would also like to thank Mrs. Joan Brown, Miss Susan Mahoney and Mr. Fred Howe for their technological assistance in histology, hematology and surgical procedures. Thanks are due to the Department of Physiology, Department of Animal Husbandry, Department of Pathology, and the De- partment of Food Science for the providing of equipment and laboratory facilities. I would like to express my indebtedness and warm appreciation to my fellow graduate students, Dr. A. A. Shaikh, Miss Shirley Johnson, Miss Masum Montakahabolayaleh, I 111 Mr. Lee Blakely, Mr. James Rae, and Miss Elli Pasteur for their friendship, assistance, and countless pleasant times spent together. I would like to thank Mrs. Alberta Neller, Mrs. Mary Tripp, Miss Julia Britten, and Mrs. Kay Butcher for their "cheerful dispositions," and warm friendship. My sincere thanks to my parents for their encourage- ment, council and assistance in a variety of ways that made this study possible. iv TABLE OF CONTENTS ACKNOWLEDGMENTS LIST OF TABLES LIST OF GRAPHS LIST OF PLATES INTRODUCTION Historical Review. Endometritis Hormones and Genital Tract Defense Tissue Leucocytogenesis. METHODS AND PROCEDURES. Preparation of Rabbits Bacterial Preparations Bacterial Inoculation Recovery of Bacteria. Histological Procedures. RESULTS. Bactericidal Activity of the Reproductive Tract. . Cyclic Estrogen in Constant Estrous Rabbits The Inflammatory Response . . . . Leucocytes of Tissue and Blood DISCUSSION. SUMMARY. BIBLIOGRAPHY Page iii vi vii viii 12 14 18 21 21 22 23 2A 32 32 39 A8 59 86 89 Table 1. LIST OF TABLES Page A Comparison Among Estrous, Pseudopregnant and Ovariectomized Rabbits, Involving Leuco- cyte Migration into the Uterine Endometrium and Lumen, Prior to and Following Bacterial Infection, with Comparative Uterine ' Bactericidal Activity . . . . . . . . 51 Effects of Inoculation of Bacterial Antigen on Cell Types Present in Uterine Genital Tissues . . . 52 vi Graph 1. LIST OF GRAPHS Page Basophilic Preplasma Cells and Globular Leuco- cytes in Estrous and Pseudopregnant Rabbit Uteri Prior To and During Induced Bacterial Infection . . . . . . . . . . . . . 53 vii LIST OF PLATES Plate Page 1. Modification of Leucocyte Numbers in Estrous and PseudOpregnant Uteri Prior To and Following Bacterial Infection . . . . . . SA 2. Tissue Leucocytes in the Uterus . . . . . 56 viii INTRODUCTION Historical Review Inflammation is an event which has captured man's imagination since earliest times. One of the classic con- cepts contributing to the understanding of inflammation was made by Celsus (25 B.C.—A5 A.D.). In Book III, from his treatise on De Re Medicina, comes the definition: "Notae vero inflammationis sunt quatuor, rubor et tumor, cum calore et dolore." These four words, redness and swelling with heat and pain define the "cardinal signs" of inflammation. With the early realization of this concept, it is sur- prising that the next maJor contribution was not made until almost 1700 years later by the English surgeon, John Hunter (1728 - 1793). He suggested that inflammation is the re- sponse to any injury and its function is to restore the af- fected part to its natural state. Hunter stated that "If inflammation develops, regardless of the cause, still it is almost the same type everywhere, because everywhere it is an effect whose purpose it is to restore the parts to their natural function." In later years Hunter taught that inflammation was primarily an increase in vessel size com- bined with the formation of new vessels as stated. He suggested "the act of inflammation would appear to be an increase action of the vessels," in addition, "the part in- flamed becomes to appearance more vascular than in the natural state, and it is probable that it is really so both from new vessels being set up in the inflamed part." These concepts remained relatively unchanged until the microscope was discovered. Virchow (1855) recognized that all events whether physiological or pathological were contingent upon the activities of the body cells. It was also his view that sequential alteration of tissue cells was the factor responsible for the entire process of inflam- mation. VI In the middle of the 19th century, Cohnheim (1877) in- vestigated the factors involved in inflammation and demon- strated conclusively that the vascular system played an essential role in the regulation of tissue temperature, red— ness, and swelling. By continuous microscopic observation following stimulation with hot wire or drops of acid on the web of a frog's foot, Cohnheim observed the following vascu- lar alterations: 1. "Sometimes a very brief constriction of the arterioles . . . probably the direct effect of the injury rather than the reaction to it." 2. "Dilation, first of the arteries, later of the veins, and in perhaps an hour a dilation of the capillaries." 3. "Acceleration of the stream in the dilated vessels, the axial stream becoming very marked." 4. "This is followed by a gradual slowing of the stream until the actual current, which has broadened out more and more, is ultimately lost. Meanwhile, the leucocytes appear to have become sticky, and tend to adhere to one another and to vessel walls in the part affected." 5. "This process and the progressive slowing of the stream may end in complete stasis, though it does not necessarily do so." 6. "At the same time an exudation of lymph is occur- ring in the vessels" and . . 7. "Diapedesis, leading to so called 'margination' of the smaller veins especially, with leucocytes, many of which gradually pass out to reach the in- jured surface, covering it eventually with a thick layer which dries to form the scab." (Johnheim further observed that circulation eventually returns followed by regeneration after twenty-four hours at the PeIPiphery of the wound. This observation resulted in the foxrmation of a theory which suggested that the inflam- matOrY’ process must be due to molecular alterations of Vessel.'walls. IIn 1883, Metchnikoff examined the effects of foreign materials on lower forms of life. In one experiment he .L: introduced rose thorns under the skin of starfish larva and observed the migration of mobile mesodermal elements around the thorns. As a result of this experiment he gave the name "phagocyte," meaning devouring cells, to the mesodermal ele— ments. However, it was not until 1891 that Metchnikoff, in his treatise on the comparative pathology of inflammation, proposed the idea of inflammation as being the reaction of wandering mesodermal cells against the foreign irritant. "Inflammation must be regarded as a phagocytic reaction on the part of the organism against irritants. This reaction is carried out by the mobile phagocytes, sometimes alone, sometimes with the aid of the vascular phagocytes or of the nervous system." As Metchnikoff elucidated the part of phagocytes in response to inflammation, he recognized two general types, the macrophages and microphages. The micro- phages included the granular polymorphonuclear leucocytes while the macrophages were divided into two subgroups; the first consisting of lymphocytes which in turn formed the large mononuclears found in the blood, while the second group, the histogenous macrOphages consist of phagocytic cells of the spleen, peritoneum and fixed tissue cells. On considering the part that these cells played in an infection, Metchnikoff (1883) stated: "In mammals the real scavengers, that is to say the phagocytes which carry out the work of absorption, are the macrophages, especially the mononuclear leucocytes. These cells play an important part in chronic inflammations such as tuberculosis, while in acute _. I inflammations, it is mainly the microphages, or neutrophile polynuclear leucocytes, that are involved." Following the reports of Metchnikoff, emphasis was placed on understanding the mechanisms involved in phagocytosis. The discoveries of Hunter, who visualized inflammation as a defensive reaction; those of Cohnheim, who considered it to be primarily a vascular reaction; and Metchnikoff's theory of phagocytosis, provided the basic foundation for the current understanding of inflammation. The first contribution which related the various cells of inflammation to a time interval following the onset of inflammation was made by Borrel (1893). In 1897, Durham ob- served the cellular changes resulting from peritoneal in— fections of guinea pigs. By immediately removing the exudate from the infected animal and examination in a hanging drop, supplemented with stained smears and tissues, Durham observed the following sequence: 1. "Stage before leucopenia" which includes a "balling together . . . of the hyaline and mexoxyphil cells (granular leucocytes), the lymphocytes being exempt" observed within two to five minutes. 2. Following this was the "leucopenic stage" which lasted about one hour and was characterized by a disappearance of cells with an increase in peritoneal fluids. 3. The "microxyphil stage" which is marked by the infiltration of a cell type referred to as polynuclear by Metchnikoff. Durham noted that the granules of these cells were similar to those corpuscles found in the blood. This stage reaches its peak at about fifteen hours. A. At this time another cell type marks the beginning of the "macrophage stage." These cells were ob- served to "actively eat" the microphils as well as bacteria and other particles. 5. Between the fourth and the sixth day the peritoneal fluids return to normal with the gradual disap- pearance of macrOphages. Durham's contribution provided the guidelines for under- standing the inflammatory sequence. However, it remained for Beattie (1903) to extend the observations of the inflammatory sequence as well as to express the results in terminology which had become more widely accepted in the short span of ten years following Durham's contribution. Commenting on the observations of Beattie, Menkin (19A0) stated: "This sequence is true of the majority of the inflammatory reactions caused either by bacteria or chemical irritants." Beattie's findings resulted in a clear understanding of the cellular sequence of inflammation as classically presented in modern pathology texts with only minor modifications. For this reason, Beattie's ob- servations will be carefully considered. Evidence for a logi— cal sequence was obtained by inducting peritonitis from in- jections of tubercle bacilli into the peritoneal cavity of guinea pigs. Smears were taken and stained from intervals of five minutes to seven days. The following is a summary of the results obtained. 1. Ten minutes after inoculation. In the presence of numerous free bacilli there were only a few polymorphonuclear leucocytes. Course granular eosinophilic leucocytes and lymphocytes were present but only in small numbers. At twenty minutes there was still no visible in- crease in polymorphonuclear leucocytes, eosino- philic leucocytes and mononucleated cells. How- ever, it is interesting to note that mitosis of mononuclear cells was observed as well as phago- cytosis of other cells by the mononuclears. Between forty minutes and one and one-half hours following inoculation, a total decrease in all cells present occurred. This is comparable to Durham's (1897) leucopenic stage. At two hours, the overall cell population was still low but at this time two types of mono- nuclear cells were noted by Beattie. The first was quite small and resembled the lymphocyte while the second morphologically resembled endothelial cells. During the interval of two and a half to four hours there was a substantial increase in polymorpho- nuclear leucocytes with a few containing bacteria. Also observed at this time were intermediates 10. between the "lymphocyte like" and the large "endothelial like" cells. With the four hour infection, the polymorpho— nuclear cells were greatly increased and seen to contain many bacilli. The proportion of mono- nuclears continued to remain low. Between six and twelve hours, the polymorphs were very numerous. At this time only a few of the bacilli were found free in the exudate. Mono- nuclears became progressively more numerous and aside from showing pronounced variations in size also exhibited nuclei which had an "eccentric or often kidney bean shape." Small masses of proto— plasm with and without nuclei, as well as being smaller than lymphocytes, were believed by Beattie to be "buds or pseudopodia" which have separated from the polymorphonuclear leucocytes. At eighteen hours the polymorphs were found to be present in the same numbers as six to twelve hours. Close examination showed that a few of the larger mononuclears contained polymorphs or their remnants in the "digestive vacuoles" of these cells. Between twenty-four and thirty hours the picture remained the same as described at eighteen hours. At thirty hours the larger forms of mononucleated cells were greatly increased and there was suffi- cient evidence of phagocytosis of polymorphs, with no bacilli being found in the exudate. ll. 12. 13. 14. From thirty-six to forty—two hours the polymorpho— nuclear cells decreased progressively in non fatal cases while medium sized mononucleated cells actively participated in phagocytosis. For the first time at forty-eight hours, the number of mononucleated cells were seen to equal and some— times surpass the numbers of polymorphs present. Mononucleated cells greatly outnumbered the poly- morphs present at fifty hours, even though a gradual decrease in mononucleated cells was occur— ring. Between seventy-two and ninety-two hours all non- fatal cases showed a continuous decrease in the cells present. Polymorphonuclear leucocytes were markedly decreased. A slight rise in coursely granular eosinophil cells was noted at this time. Mention of this was not made earlier due to the low levels observed. Mononucleated cells could be seen undergoing continuous mitosis. Beattie's observations of cellular sequence remain as guideposts for future analysis of inflammatory reactions. It should be noted, however, that no direct mention of cellu— lar migration to an inflamed area was made. Leber (1888) introduced the-concept of chemotaxis following infection of the cornea with Aspergillus. It was reported that leucocytes were attracted by the toxic sub- stances, called phlogosin, and of bacterial origin. Massart 10 and Bordet (1891) demonstrated that disintegrating leuco— cytes formed a product which was able to attract leucocytes in tissues. Mallory (1918), in his text on pathologic histology, stated that the order of blood cell infiltration is-the same regardless of the cause of inflammation. Poly- morphonuclear leucocytes are the first to imigrate into the inflamed area with monocytes arriving shortly thereafter. Lymphocytes become abundant after the inflammation has lasted several days and often remain until after the lesionhas healed. More recently, Menkin (1940, 1956) was able to ex- tract a factor from inflammatory exudates which he called "leukotaxin." It was suggested that this factor was liber— ated from the tissues and increased the permeability of the capillary wall and directed the migration of granulocytes. Supporting this theory were the clumps of leucocytes in capillaries which contained "leukotaxin" crystals as well as a granulocyte accumulation around these crystals. It was also reported that neutrophils are attracted in large numbers to pus cocci as well as other microorganisms and certain chemicals (Nonidez and Windle, 1953). Trypsin and porteases found in macrophages (histiocytes) apparently empower these cells to digest engulfed proteins (McCutcheon, 1953). Closely related to this ingestion is the death of the macrophages and the release of various constituents for reuse by other cells. Therefore, phagocytes not only contribute to the acute phase by protecting the body but, in addition, contribute to the subsequent repair process (Best and Taylor, 1943). The ll histogenous wandering cells appear in the inflammatory area later than the polymorphonuclear cells. They are very mobile and, in general, phagocytize cells and cell debris rather than bacteria, as a preparatory step for eventual tissue regeneration (Evans, 1947). As mentioned earlier, eosinophils are present through- out the inflammatory period with lowest numbers observed in the acute stages. For this reason the eosinophil has been associated with the convalescent phase of an infection. In certain cases, they may be abundant in chronic inflammation, an example being in the skin (Braunsteiner and Zucker- Franklin, 1962). Beyond this, very little is known about the functions of the eosinophil. Another cell type associ- ated with chronic infection and the late stages of acute in- fection is the lymphocyte. Rich (1936) discussed the inflam- matory process and observed that lymphocytes were "congre- gated often in the more peripheral parts of the lesion . and have the appearance of phlegmatic spectators passively watching the turbulent activities of the phagocytes." More recently Kelsall and Crabb (1959) summarized lymphocytic participation at the inflammatory site by reporting that they have "the function of storing and supplying nucleic acids, adenosine triphosphate, histones, certain enzymes and other proteins and certain minerals for use by the proliferating cells in the healing, or last stage, of the subchronic or chronic inflammatory process." Lymphocytes are also reported 12 to be transporters of antibodies manufactured by reticu- loendothelial cells and plasma cells. The plasma cell, whose function was greatly debated until very recently, is also found in great numbers in the chronic infection. With the develOpment of the flourescent antibody method by Coons 33:11. (1955) conclusive evidence was reported on the presence of specific antibodies within the cytoplasm and nuclei of the plasma cells. Endometritis In considering the defensive reaction to infection, one should not overlook the interesting condition of endo— metritis and its relationship to infertility. Awareness of this condition was apparent in Switzerland in 1843 when the suggested therapy for endometritis was a creasote soaked plug inserted into the cervix, as reported in the Schweizer Archiv fur Turheilkunde (Dawson, 1950). The work of Zachokke, published in 1900 and quoted by Albrechtsen (1917) stated that there was a relationship be- tween endometritis and prolonged anestrus and he observed a condition of persistant corpora lutea during prolonged anestrus. Wapf. Hess (1905) further clarified the "utero- tropic" effect of the corpora by stating that the uterine disease was secondary to the primarycondition of ovarian abnormality. Sharp conflict arose when Albrechtsen (1917) asserted that endometritis eventuated from primary microbial infections which established themselves in the uterus. 13 Furthermore, he stated that ovarian involvement stemmed from uterine infection which spread to the ovaries. Signifi- cant emphasis was placed for the next twenty years on dis- infection of the reproductive tract as the mode of treat- ment for endometritis. Variations of this treatment included uterine irrigation with weak solutions of iodine and copper sulphate. Positive clinical results supported Albrechtsen's theory that bacteria were the primary cause of infection. Neilsen (1926) proposed a slight modification of Albrechtsen's theory after re-evaluating the relationship of the ovary to the primary infection and concluded that ovarian manipulation was comparable to uterine irrigation from the therapeutic standpoint, that is, modification of the functional activity of the ovary changes uterine sensitivity to the same bacteria. Following the discovery of endocrine control of repro— ductive physiology, Riedel (1935) reported that treatments involving bactericidal agents had their value as irritants and not disinfectants. Finally, Hignett (1940) and Wille (1942), demonstrated that uterine stimulation could evoke an ovarian response which altered the reproductive system. At the same time, Rowson and_Spriggs (1942) reported that treatment must be directed to the ovary so that hormonal changes might alter uterine functional activity and hence resistance to bacteria. Albrechtsen (1917) and Nielsen (1949) presented evi— dence that endometritis is due to uterine infection which l4 eventuates in cattle infertility. More recent articles have substantiated this position. Hormones and Genital Tract Defense It was reported by Quinlan (1929) that uterine in- fection is the main cause of post-partum infertility, as many problems arise following parturition when the cervix remains open. Hormonal influence on infertility was recog- nized by Marshall and Hammond (1932) when it was reported that per rectum nidation of the corpus luteum often im— proved fertility. It was suggested that removal of the corpora provided one method by which the estrogen—proges- terone ratio could be altered and bactericidal activity enhanced. Laing and Day (1947) observed that heifers bred to bulls carrying Trichomas foetus infection invariably aborted. If these heifers were rebred to a clean bull, the majority again aborted. However, if cows were permitted to go through a normal estrous cycle after the first abortion, before being bred to a clean bull, the majority made a spontaneous recovery, conceived, and delivered normally. It is interesting to note that the authors reasoned that the intervening follicular phase and estrogen production modified the uterine media in such a way that it would not support the infection, while progestational influence pro- vided a uterine media more conducive to bacterial repro- duction. 15 Black gt_a1. (1951, 1953a, 1953b), McDonald e§_al. (1952) and Rowson gt_a1. (1953), demonstrated that the uteri of follicular phase and luteal phase animals differ in their inflammatory response to infectious agents. The uteri of luteal phase rabbits and cows respond to inocu- lation with bacteria or bacterially contaminated semen by mass leucocytic infiltration of the endometrium and accumu— lation of pus in the lumen one to two days postinoculation. It was also reported that inoculated uteri of estrous animals exhibited relatively little gross evidence of in- fection, with only moderately increased numbers of leuco- cytes either in the tissue or uterine exudate. The greater inflammatory response of luteal phase animals did not re- flect a.more effective defense mechanism. The uteri of follicular phase rabbits and cows contained fewer remaining organisms at intervals of four and forty-eight hours, while large numbers of bacteria were recovered for the same time interval from the uterine exudate of luteal phase females. These results were typical of all animals investigated by Black M- (1951:) and Hawk M. (1955, 1957). It was demonstrated by Hawk gt_al. (1957) that the difference in numbers of bacteria recovered from inoculated uteri of follicular and luteal phase animals was apparently due to the greater bactericidal activity of the uteri of estrous animals, since the difference could not be attributed to altered nutritive properties of the respective uteri. Furthermore, Hawk (1958) observed that the exudate from 16 inoculated uteri of estrous rabbits contained substances capable of high bactericidal activity at four hours post- inoculation, after the leucocytes were removed. Pseudo- pregnant rabbit exudate contained less of the active factors at the same time interval. The leucocytic response and the accumulation of noncellular factors were associated with a high degree of bactericidal activity in estrous rabbits at four hours. In pseudopregnant rabbits the leucocytic re- sponse was considerably slower and cell free exudates con- tained substances which were only slightly bactericidal. Rowson g£_al. (1953) and Black gt_al. (1954) re- ported that twenty-four hours postinoculation uteri of luteal phase animals contained greater quantities of viable bacteria than the uteri of follicular phase animals, even though they contained more pus at this time. Hawk gt_a1. (1960b) demonstrated that bactericidal activity was grossly proportional to the number of leuco- cytes in the uterine lumen regardless of the endocrine state. However, there were more Escherichia coli alive at sixteen hours in the pseudopregnant uteri when there were large numbers of leucocytes, suggesting some basic differ- ence in uterine defense mechanism. These differences, as stated by Hawk may be "a partial inhibition of the defensive elements present in the pseudOpregnant rabbit uteri or the action of defensive elements peculiar to the estrous rabbit." Broome gt_al. (1960), Hawk §£_al. (1961), and Brins- field et a1. (1963) suggested that endogenous or exogenous 17 progestin retards the leucocytic response and bactericidal activity in rabbit and sheep uteri. When estrogen was ad- ministered to ovariectomized rabbits the leucocytic response was stimulated while progesterone inhibited the influx of leucocytes. When ovariectomized does were compared with ovariectomized progesterone rabbits, leucocytes accumulated more rapidly in the progesterone treated endometrium but migrated into the lumen more slowly at four hours than in its untreated counterpart. However, the influence of the hor- mones on the leucocytic response could not be accounted for by variation in the normal leucocytic infiltration of the endometrium (Hawk g£_a1., 1960a). Brinsfield et_a1. (1964) reported that progestin de— lays the leucocytic response and bactericidal activity in sheep uteri. Hecter gt_al. (1941, 1942) examined the influence of ovarian hormones on uterine vascular permeability in normal rabbits. Estrogen and progesterone increased the uptake of trypan blue dye by the uterus. Hawk et_a1. (1963a) recog— nized an increase in permeability in estrous and ovari- ectomized estradiol treated rabbits beyond that of pseudo- pregnant or ovariectomized progesterone treated rabbits. Burke and Miles (1958) stated that leucocytic emigration oc— curs independently of increased vascular permeability. This was later confirmed and expanded by Hurley and Spector (1961). It is not surprising therefore, that the pseudopregnant rabbit develops an intense endometrial leucocytosis after four hours with only slightly increased permeability. 18 Tissue Leucogytogenesis Stockard and Papanicoloau (1917) examined the relation- ship between the vagina, uterus and ovaries using a vaginal smearing technique developed by Papanicoloau in 1916. The perfection of this technique led to the recognition by Papanicoloau (1933) of a sexual cycle in women. The cycle is characterized by the presence of leucocytes in the vagi- nal smear during the luteal phase of ovarian activity in the guinea pig (Stockard and Papanicoloau, 1917), rat (Long and Evans, 1922), and human (Pouchet, 1847; O'Leary, 1929). The follicular phase gives rise to a vaginal smear devoid of leucocytes with the simultaneous accumulation of eosino- philic and neutrophilic leucocytes in the endometrium of the uterus and vagina. It was suggested (Papanicoloau and Shorr, 1936; De Allende and Orias, 1950; and Edgren, 1959) that an epithelial barrier is developed at estrus which inhibits leucocytic migration into the lumen. Fluhamann (1932), reported the disappearance of poly- morphonuclear leucocytes from the vaginal smear of ovari— ectomized rabbits following estrogenic treatment and simul- taneous leucocytic accumulation in the endometrium of the uterus and vagina. This was confirmed in cattle, sheep, and swine by Nellor (1962) and in guinea pigs by Montakahabolayaleh (1964). In contrast, progestational treatment resulted in an increase in polymorphonuclear leucocyte infiltration into the vaginal tissue and smears of ovariectomized or menopausal 19 women (De Allende and Orias, 1950) normal cycling rats (Selye gt_al., 1935), mice (Beaver, 1960), and estrogen treated ovariectomized rats (Edgren, 1960). Nellor (1963, 1965a) reported the presence of leuco- cyte-like cells arising from connective tissue cells of the oviduct. An increased mobilization of "lymphoblast like cells" was noted during the luteal phase of the estrous cycle or following progestational treatment. Migration of these cells and their eventual extrusion into the lumen oc- curred during progestational treatment in the intact luteal phase female as well as in ovariectomized progestin treated females. Estrogen treatment alone did not maintain migration or penetration of "lymphoblast like cells" into the lumen, but estrogenic treatment during the luteal phase or given simultaneously with progestin induced a morphological re- organization of the wandering cells. This was characterized by an "eosinophilic granulation of the cytoplasm, followed by nuclear chromatin fragmentation within the cell." This process terminated in the destruction of the cell. In the same year Nellor and Brown (1965b) reported the presence of leucocyte-like cells in the genital tissues land smears of cattle, sheep and swine which are not of im- mediate blood origin. It was suggested that these cells arise from undifferentiated mesenchymal cells and develop into plasma cells during the luteal phase of the cycle. These cells are then capable of migration through the stroma and into the lumen. As the luteal phase wanes, follicular 20 phase estrogens initiate nuclear membrane fragmentation and degranulation of the wandering plasma cells, resulting in degenerating cells resembling polymorphonuclear leuco- cytes within the genital tissues. The current research program was designed to examine the relationship between hormonal influence and the genital tract defense system. The primary difference between prior studies and the current study on hormone influence of geni- tal tract bactericidal activity is that in this research program emphasis has been placed on establishing the role of tissue leucocytogenesis in the acute inflammatory re— sponse. METHODS AND PROCEDURES Preparation of Rabbits In the present study, sexually mature, female, New Zealand rabbits between eight to twelve months of age were used. Does weighing three to four kg. were maintained in isolated cages with food and water ad lib. The experiments were designed in order that three endocrine states were represented in the rabbits. 1. Estrous rabbits were held in isolated cages for a minimum of one month prior to use. This pre- caution was made to provide estrous females since transporation or altered environmental conditions are known to induce pseudopregnancy in rabbits. Pseudopregnancy was induced by the injection of 100 I.U. of Pregnant Mare's Serum (PMS) and 100 I.U. Human Chorionic Gonadotropin (HCG) into the marginal ear vein of estrous rabbits which were then used nine to ten days later. Ovariectomized rabbits were prepared by a para- lumbar incision and used fourteen to twenty-one days later. 21 22 The following numbers of rabbits were included in the current study. a. Estrous rabbits 20 b. Psuedopregnant rabbits 6 c. Ovariectomized rabbits 8 d. Estrous rabbits (incon- clusive results) The latter category, estrous rabbits yielding incon- clusive results, represents treatments directed at tagging blood leucocytes, both in vivo and in vitro, with india ink, iron filings, trypan blue, and killed bacteria. Bacterial Preparations The bacteria used in this study were Escherichia coli. Bacterial suspensions were prepared by culturing for twenty- four hours in trypticase soy broth. Following incubation at 37° C., the E. coli were separated using an international clinical centrifuge at approximately 4000 rpm. for twenty minutes. Following decantation of the broth, the bacterial pellet was resuspended in sterile 0.90% saline and diluted to a standard turbidity using a Coleman Junior Spectrophoto- meter. Bacterial concentrations ranged from 2.7 — 8.0 x 108 E. coli per 0.2 ml. of inoculated suspension. In addition to turbidity measurements, samples of the culture were quanti— fied using plate counts to determine the number of viable bacteria. Within one hour of inoculation serial ten fold dilutions were made of the suspension, using sterile 0.90% 23 saline. Preheated trypticase soy agar was added to sterile petri dishes containing one ml. suspensions of 10-5, 10-6, ‘8, 10-9 dilution. Following stirring, the plates 10'7, 10 were inverted and incubated for twenty-four to forty-eight hours and then counted. Bacterial Inoculation Rabbits were anesthetized with six per cent sodium pentobarbital solution, consisting of 10 m1. alcohol and 90 ml. of distilled water, the injections being made into the marginal ear vein. After clipping abdominal hair, a midline incision was made exposing the uteri. A ligature was placed around the tubo-uterine junction of the right uterine horn with care being taken not to occlude the lateral arcuate vein. In addition to this an incision was made in the anterior vagina and a purse-string suture was placed on the cervix of the horn to be inoculated with bacteria. This resulted in the formation of a pouch capable of holding the bacterial inoculum. The inoculum was injected through the cervix into the pouch with a twenty-six gauge needle. All infected rabbits were inoculated with a 0.2 ml. saline sus- ‘pension containing approximately 3.7 x 108 cells. Recovery of Bacteria Following introduction of bacteria, animals were main— tained in an anesthetized state until the experiment was terminated by an overdose of sodium pentobarbital. The uteri, 24 ovaries, fallopian tubes, cervix and vagina were then re— moved. Excessive fat was dissected from the ligated uterine horn containing bacteria and inflammatory exudate. The liga- ture at the tubo-uterine junction was released to permit ac— cumulated fluids to flow into the sterile test tube. At the same time, the horn was flushed with 5-10 m1. of trypticase soy broth to assure total recovery of bacteria and leucocytes. Two one ml. samples of the inflammatroy exudate were then serially diluted and plated according to the procedure used in preparing the inoculum in order that the surviving E. coli could be determined. In addition, smears of the exudate were made as well as total leucocytic counts, by means of a hemo— cytometer. Tissue sections of control and treated uterine horns were prepared for subsequent histological analysis. Since preliminary studies indicated a positive but de- layed bactericidal response by the contralateral control horn, it was necessary to surgically remove a small section of the control uterine horn prior to infection of the remaining horn in order that a truer histological comparison might be made. Histological Procedures Tissues sections were removed from the middle portion of the uterine horns and fixed in Carnoy's solution (10% acetic acid, 30% chloroform, and 60% absolute alcohol), ab- solute alcohol, acetone, ten per cent formalin, and Carnoy's solution without acetic acid. Specimens in absolute alcohol were processed by the autotechnicon after fourteen to 25 sixteen hours of fixation, while the remaining tissues were processed two to five days later. Following paraffin im- bedding, sections were cut at six micra on the microtome. Samples of uterine inflammatory exudate and systemic blood from the central ear vein or heart, were smeared on glass slides immediately after recovery. Smears were then fixed, within fifteen to twenty seconds, in the tissue fixa- tives previously mentioned and processed by the autotechicon. The only exceptions were duplicate air dried slides. In addition to tissue sections and smears being processed by the autotechnicon, additional specimens were frozen and in the case of tissues, sectioned at thirty to forty micra using a cryostat. Tissues and smears were then stained with per- oxidase, Wright's and Giemsa's stains. Stains employed on autotechnicon processed tissues and smears were Papanicoloau's, Wright's, Alcian blue, toluidine blue, methylene blue, hematoxylin and eosin, and Wright's stain modified for tissues. The Papanicoloau stain used was a modification by Montakahabolayaleh (1964). This resulted in a clearer differentiation of nuclear and cytoplasmic components when applied to genital tract tissue staining. The nuclei stained with hematoxylin while glycogen and mucin in the cytoplasm were accentuated by Orange G..EA-65, which is a combination of light green, bismark brown, and eosin, intensifies the acidophilia as well as the glycogen and mucin found in the cytoplasm. The procedure for Papanicoloau staining in this study is as follows: 26 Tissue (time) Smear (time) r 12. 13. 14. 15. l6. l7. l8. 19. 20. 21. 22. 23. OKOCIDNQU'I Xylene Xylene Xylene 100% alcohol 95% alcohol 80% alcohol 70% alcohol 50% alcohol 30% alcohol Distilled water Harris hematoxylin Distilled water Distilled water Distilled water 50% alcohol 70% alcohol + NHuOH 70% alcohol 80% alcohol 95% alcohol Orange G 95% alcohol 95% alcohol 95% alcohol 5 minutes 5 minutes 5 minutes Rinse II II 1 1/2 minutes Rinse II II 3 minutes Rinse II II 5 minutes 5 minutes 5 minutes Rinse 1 minute Rinse II II II 1 1/2 minutes Rinse II II 1 1/2 minutes Rinse II II 27 Tissue (time) Smear (time) 24. 25. 26. 27. 28. 29. 30. 31. 32. E.A“65 95% alcohol 95% alcohol 95% alcohol 100% alcohol 50% Xylene + 50% Absolute Alcohol Xylene Xylene Permount 5 minutes Rinse 3 minutes 3 minutes 3 minutes 3 minutes Rinse II II II 3 minutes 3 minutes 3 minutes blood smears which were air dried. Wright's stain (1902) was used on uterine exudate and Blood and exudate smears were flooded with Wright's stain for one minute and then washed with an equal amount of buffer (0.7 gm. monobasic sodium phosphate and 0.3 gm. dibasic sodium phosphate per liter, pH-6.5) for five minutes followed by a distilled water rinse. The Alcian blue method adopted was a modification of .Mowry (1956) for staining acid mucopolysaccharides and was utilized at a pH of 2.5 to 3.0. as follows. Xylene - 5 minutes Xylene — 5 minutes- Xylene - 5 minutes The staining sequence is 28 4. 100% alcohol - rinse 95% alcohol - rinse 80% alcohol - rinse 5 6 7. 70% alcohol - rinse 8 50% alcohol - rinse 9 30% alcohol - rinse 0 Distilled water -.rinse 11. 33% Acetic Acid - rinse 12. 0.1% Alcian blue in 3% acetic acid* - 30 minutes 13. Tap water - rinse 14. Distilled water - rinse 15. Harris hematoxylin - 1 minute 16. Distilled water 17. Distilled water 18. Distilled water 19. 30% alcohol — rinse 20. 50% alcohol - rinse 21. 70% alcohol - rinse 22. 80% alcohol - rinse 23. 95% alcohol — rinse 24. 100% alcohol - rinse 25. 50% absolute and 50% xylene - 3.minutes 26. Xylene - 3 minutes 27. Xylene - 3 minutes 28. Permount *Alcian blue solution - 0.1% Alcian blue in 3% glacial acetic acid, pH-2.5 to 3.0. Filter, add thymol crystal. 29 Toluidine blue was utilized to demonstrate the presence of metachromatic granules at a pH below 0.5. This stain was used to define the presence of the tissue leucocytogenic series stem cells described by Nellor (1965b). The procedure used is: 1. Xylene - 5 minutes 2. Xylene - 5 minutes 3. Xylene — 3 minutes 100% alcohol - rinse 5 95% alcohol - rinse 6 80% alcohol - rinse 7. 70% alcohol - rinse 8 60% alcohol - rinse 9 Toluidine blue - l to 2 minutes* 10 Rapid tap water - rinse 11. Acetone - 3 minutes 12. Acetone - 3 minutes 13. Xylene — 3 minutes 14. Xylene - 3 minutes 15. Permount The Harris hematoxylin stain was used as described by Mallory (1944) with the counterstain of eosin prepared ac- cording to Humason (1962). Wright's staining was modified in this study for utilization with tissue sections in order that smear and tissue leucocytes could be compared following identical processing. The candidate is not aware of any *0.2% toluidine blue in 60% alcohol. pH-0.1. 30 reports appearing in the literature demonstrating the appli— cation of this stain to tissue sections as well as smears which have been prepared in fixative and autotechnican.processed. 1. Xylene — 5 minutes 2. Xylene - 5 minutes 3. Xylene - 5 minutes 4. 100% alcohol — rinse 5. 95% alcohol - rinse 6. 80% alcohol - rinse 7. 70% alcohol - rinse 8. 50% alcohol — rinse 9. 30% alcohol — rinse 10. Distilled water - rinse 11. Wright's stain* — 2 minutes 12. Equal part buffer** — 7 minutes 13. Distilled water — rinse l4. Distilled water - 3 minutes 15. 95% acetone and 5% Xylene - rinse 16. 70% acetone and 30% Xylene - rinse 17. 50% acetone and 50% Xylene — rinse 18. 30% acetone and 70% Xylene - rinse l9. 5% acetone and 95% Xylene - rinse *Wright's stain - 0.1 gm. stain per 60 cc. redis- tilled methyl alcohol. **Buffer - 0.7 gm. monobasic sodium phosphate 0.3 gm. dibasic sodium phosphate pH = 6.5. } per liter distilled water 31 20. Xylene - 3 minutes 21. Xylene — 3 minutes 22. Permount RESULTS Bactericidal Activity of the Reproductive Tract Interest in the primary stages of acute infection in rabbit uteri resulted in the concentration of the research being placed on the first four hours following bacterial inoculation. Bacterial suspensions were inoculated into uterine horns and inflammatory exudates and genital tissues representing one, and two, and four hours were studied. The superior defense mechanism in infected uteri of estrous rabbits is indicated by the onset of bactericidal activity at two hours and the eventual distruction of ninety- nine per cent of the introduced bacteria within four hours following inoculation. Within this same time, the uteri of pseudopregnant rabbits, with comparable amounts of E. coli introduced, destroy only twenty to thirty per cent of the inoculated bacteria. Ovariectomized rabbits exhibit a higher uterine bactericidal activity than the pseudopregnant females at the same infection intervals, but less activity than estrous rabbits (Table l). The fluid recovered from the infected horn was at a minimum during the four hours following inoculation in estrous, ovariectomized, and pseudopregnant females. It was therefore 32 33 necessary to aseptically flush the infected horn to assure recovery of the exudate. The ligated inoculated horn was injected with 10 m1. of trypticase soy broth with a two to three minute interval of time permitted to allow homogeneous equilibration. Samples were then obtained in order that bacteria and leucocyte numbers could be determined. The only exception to the minimal volumes of fluid obtained, was in twenty—four hour infections where 0.75 to 1.00 ml. of exudate were usually recovered. Total leucocytes present in the tissues and lumina of infected uteri are expressed on the basis of relative influx (Table 1). In estrous female rabbits, increasing numbers of wandering cells in the uterine tissues are related to luminal influx and bactericidal activity. That is, as blood leuco- cytes and tissue leucocytes increased in number in the tissues, increasing numbers of leucocytes were observed in the infected uterine lumina. It appears, however, that.pseudopregnant fe- male rabbits have a similar, but delayed leucocytic infil— tration, since leucocytes appear in the uterine exudate for the first time at four hours postinoculation. Consideration of the ovariectomized females tissue response to antigen suggests that leucocytic migration more closely resembles the estrous than the pseudopregnant response (Table l). Histological examination of infected and noninfected estrous uteri and infected psuedopregnant rabbit uteri, demonstrated a cell series to be present that originated from an undifferentiated cell in the connective genital 34 tissue, similar to that described by Nellor and Brown (1965b) in other mammals. This cell series was apparent in the tissues following fixation in Carnoy's solution and staining with Alcian blue-hematoxylin between a pH of 2.5 to 3.0. Alcian blue staining cells were apparently under— going a series of morphological changes as they migrated from the connective tissue to the uterine lumen. Before considering the quantitative variations in uterine leucocytes, the following description of cell types of tissue origin is presented to clarify the terminology utilized throughout the remainder of this thesis. The cate- gories define the stages of a continuous series of cells in the genital tissues of control estrous and infected pseudo- pregnant rabbit. Basophilic fibroblast-like cell. (7; 8) The basophilic fibroblast-like cell is elongated and contains a spindle-shaped nucleus with dense chromatin. The cytoplasm, which has distinct and uniform granules, is Alcian blue positive (pH 2.5-3.0) methylene blue positive (pH 0.5) and burgundy in color with modified Wright's stain (pH 6.5). This cell is not associated with the endothelial system, and in the majority of the sections examined was located in the endometrium just below the uterine epithelium. Basgphilic preplasma cell. (94 10) This cell is identified by a rounding up of the nuclear chromatin resulting in a spherical or oval appearance. At 35 this stage, the nucleus is centrally placed and contains evenly dispersed chromatin. The cytoplasm contains distinct uniform granules which stain Alcian blue positive (pH 2.5- 3.0) and burgundy with modified Wright's stain (pH 6.5). This cell was scattered throughout the endometrium and occasionally seen in the myometrium. It did not appear to be associated with endothelial tissues. Basophilic lobulated preplasma cell. (11) This cell is characterized by nuclear pinching or indentation resulting in a nuclear arrangement resembling a kidney bean. Nuclear division, not dissimilar to nuclear budding, is not accompanied by cytoplasmic division. The cytoplasm contains distinct uniform granules which are Alcian blue positive (pH 2.5-3.0) and stains burgundy with modified Wright's stain (pH 6.5). During the inflammatory reaction, this cell forms multilobulated basophilic cells (12). Basophilic multilobulated cell. (12) The outstanding identifying feature of this cell is the progressive nuclear division, apparently occurring with- }out cytoplasmic division, resulting in a polychromatin cell with dissimilar sized chromatin lobules, not connected by chromatin strands. The cytOplasm, with its distinct and uniform granules, is Alcian blue positive (pH 2.5-3.0) and stains burgundy with modified Wright's stain (pH 6.5). 36 Degranulating basophilic multilobulated cell. (13) The terminal stages of morphogenesis, which precedes cellular death, results in further multilobulation of the nucleus and is accompanied by the progressive degranulation of the cytoplasm. This cell contributes to the large number of degenerative cells (20) in the inflammatory tissue and can be identified by the granules Alcian blue positive (pH 2.5-3.0) and stain burgundy with modified Wright's stain (pH 6.5) remaining in the cytoplasm or scattered in the im- mediate vicinity. Globular 1eucocyte. (14, 15) The nucleus is eccentrically located and the cytoplasm may contain a clear cytoplasmic halo located at the nuclear margin. The cytoplasm contains granules which vary in size from very fine to vesicular in nature, staining Alcian blue positive (pH 2.5-3.0) and burgundy with modified Wright's stain (pH 6.5). Mature_plasma cell. (16) The identifying feature of this cell is its charac- teristic wedge shaped nuclear chromatin, aggregating at the nuclear membrane, and suggesting a wheel with accompanying spokes. The nucleus of this cell is placed eccentrically and a cytoplasmic halo is evident with methylene blue stain- ing. The cytoplasm is uniformly nongranulated when stained with methylene blue and neutrophilic following staining with Alcian blue (pH 2.5-3.0) or hematoxylin and eosin. Staining 37 with modified Wright's stain (pH 6.5) demonstrates the presence of small quantities of granules which stain bur— gundy. Eosinophilicgplasma cell. The nucleus of this cell is eccentric and is identi- fied by a marked condensation of nuclear chromatin creating a configuration resembling a spoked wheel. The cytoplasm, which is neutrophilic with Alcian blue, is uniformly eosino- philic with hematoxylin and eosin. This cell varies con- siderably in the number of distinct eosinophilic granules present. Traces of burgundy granules are present when this cell is stained with modified Wright's stain (pH 6.5). Prefragmentation plasma cell. (17) As the eosinophilic plasma cell continues to mature large and distinct eosinophilic granules, which stain uni- formly with eosin, are apparently being released to the extra- cellular areas. The eosinophilic intensity progressively de- creases as the granules are dispersed throughout the tissues. The nucleus, with its eccentric location, contains nuclear chromatin which condenses and forms distinct rings or bands around the nuclear membrane. The cytoplasm is neutrophilic with Alcian blue but contains remnants of burgundy granules when stained with modified Wright's stain (pH 6.5). 38 Fragmenting plasma cell. (18, 19) Eosinophilic granules are not as numerous as in the prefragmenting plasma cell and do not stain as intensely with eosin. The cytoplasm stains neutrophilic with Alcian blue but does contain sparse quantities of burgundy gran- ules when stained with modified Wright's stain (pH 6.5). An obvious pinching of the chromatin rings or bands in the nucleus results in sausage shaped chromatin rings, horse- shoes and broken strands. Dissolution of the nuclear mem- brane is apparent at this stage of fragmentation. Terminal fragmenting cell. (20) The final stages of morphogenesis, which eventuates in cellular death, is proceeded by dispersion of the nuclear chromatin throughout the cell. Disruption of the nuclear membrane releases chromatin particles until numerous dis- sociated chromatin particles are evident.‘ This cell con- tains a faintly eosinophilic nongranular cytoplasm. If the cytoplasmic membrane ruptures the chromatin particles are scattered in the connective tissue. If the membrane is in- tact, the cytOplasm may be neutrophilic or basophilic with Alcian blue (pH 2.5-3.0) depending on the cell, that is, the mature plasma cell (16) or degranulating bas0philic multilobulated cell (13), from which it evolved. Residual quantities of burgundy granules are evident following stain— ing with modified Wright's stain (pH 6.5). 39 Cyclic Estrogen in Constant Estrous Rabbits Sexually mature, noninfected, control estrous rabbit genital tissues were histologically examined for evidence of cyclic estrogen influence following staining with Papanicoloau's stain, hematoxylin and eosin, and Alcian blue. Examination of vaginal and uterine tissues demon- strated the presence of distinct fluctuations in the de- velopment of stromal and epithelial components between con- stant estrous animals. That is, it became apparent that constant estrous rabbits do not exhibit a constant ovarian hormone secretory state. Vaginal epithelium cell height was therefore used to estimate the relative levels of estrogen present. Three estrous rabbits had vaginal epithelia indi- cative of high, intermediate and low levels of estrogen. The following is a description of the genital tract cytology characteristic of varying endocrine states in three estrous rabbits. Rabbit 1. (Low level of estrogen). The superficial epithelium of the vagina ranged from low squamous to stratified squamous while the uterine epithelium was predominantly columnar. The endometrium of the vagina contained large numbers of mature plasma cells (16) and eosinophilic plasma cells, with a minimum number undergoing fragmentation (18, 19). The uterus was relatively free of mature (l6) and eosinophilic plasma cells and low 40 numbers of terminal fragmenting cells (20) were evident. The majority of these cells were present in the uterine epithelium. Rabbit 2. (Intermediate level of estrogen). With an increasing level of estrogen the superficial epithelium of the vagina was predominantly medium to high stratified squamous. Evidence for a transitional estrogen- progesterone balance was further indicated by the concen- tration of mature (16) and eosinophilic plasma cells in certain areas and large numbers of fragmented cells (18, 19, 20) in others. Comparison of relative numbers of mature (16) and eosinophilic plasma cells demonstrated that the greater percentage were eosinophilic plasma cells. The uterine epithelium was tall columnar and the majority of leucocytes in the uterine connective tissues were fragment— ing (18, 19) and terminal fragmenting plasma cells (20). Rabbit 3. (High level of Estrogen). The vagina was characterized by the presence of four to five layers of highly proliferated stratified squamous cells, with a paucity of eosinophilic plasma cells and fragmenting cells (18, 19, 20) in the subepithelial layers. The presence of low numbers of mature plasma cells (16) persisted even as estrogen levels increased. The epithelium of the uterus was high cuboidal. Intermediate numbers of fragmenting (18, 19) and terminal fragmenting cells (20) 41 were dispersed throughout the uterine stroma and epithelial layers. The Inflammatory Response Estrous uteri Control noninfected, estrous rabbits.--Noninfected estrous rabbit uteri (Plate 1, Fig. 1), contain moderate numbers of all of the tissue leucocytes previously de— scribed (Table II). The majority of these cell types were evident in moderate numbers throughout the endometrium, epithelium, and infrequently in the myometrium. Examination of the connective tissues indicated that the majority of plasma cells present were eosinophilic with eosin staining. The cell types which were Alcian blue positive (pH 2.5- 3.0) and burgundy in color with modified Wright's stain (pH 6.5) included the basophilic preplasma cells (9, 10), baso- philic lobulated preplasma cells (11), basophilic multi— 1obulated cells (12), and degranulating basophilic multi- lobulated cells (13). Examination of noninfected uterine tissues of four control estrous rabbits indicated that these basophilic cells were present in basal or minimal numbers and varied only slightly between estrous animals. The epithelial cells, which contained remnants of eosinophilic terminal fragmented cells (20), varied in size from cuboidal to columnar (Plate 1, Fig. 1). Staining with Alcian blue (pH 2.5-3.0) and modified Wright's stain 42 demonstrated the presence of globular leucocytes (14, 15) the vesicles of which stained Alcian blue and burgundy positive. Examination of all noninfected estrous uteri in- dicated that basal numbers of this cell type were present (Graph 1). One hour postinoculation, estrous rabbits.--A com- parison of noninfected estrous uteri with one hour infected uteri indicated that no gross changes were apparent, that is, at 40x magnification the uterine genital tissues appear similar to noninfected estrous control uteri (Plate 1, Fig. 1). However, at 100x oil immersion, a marked increase in mature (16), eosinophilic and fragmenting (18, 19) plasma cells was noted. However, eosinophilic plasma cells appeared to increase more rapidly relative to mature (16) or frag- menting (18, 19) plasma cells, even though all cell types were more numerous. In considering the basophilic cell types present in the genital tract tissues, the only marked alteration was the apparent increase of globular leucocytes (l4, 15). This cell type increased seven times over the total number of globular leucocytes present in the noninfected control tissues (Graph 1). A slight increase in vascularity was noted when the one hour infected estrous rabbit was compared to noninfected control uteri. However, in contrast to this, there was no noticeable increase in blood polymorphonuclear leucocytes in the major vessels supplying the uterus. 43 Two hours postinoculation, estrous rabbits.--At two hours there was a marked increase in horseshoe shaped frag- menting plasma cells (l8, 19), (Plate 1, Fig. 2). Accompany- ing this rise was a significant increase in prefragmenting plasma cells (17) and mature plasma cells (16). These cell types, along with their characteristic morphological fea— tures, contained eosinophilic granules which were burgundy negative with modified Wright's stain and Alcian blue nega- tive. In contrast to the earlier infection of one hour, the number of eosinophilic plasma cells remained unchanged even as mature plasma cells (16) and prefragmenting plasma cells (17) increased. It is interesting to note that the only baSOphilic cell to increase at this two hour period was the degranu- lated basophilic multilobulated cell (13). This was indi- cated by the number of cells containing burgundy granules. All other cell types remained numerically constant. These findings indicated that a rapid turnover of basophilic pre- plasma cells (9, 10) and basophilic lobulated preplasma cells (11) was taking place, since the total number of basophilic preplasma cells (9, 10) remained at levels similar to non- infected controls. The only other cell type which was altered was the globular leucocyte (14, 15) which declined in number by this time (Graph 1). The vascularity of the two hour infected uteri was similar to that seen in the one hour infection. However, 44 extravascular red blood cells were evident for the first_ time in the endometrium. Peripheral blood leucocytes in the outer vessels of infected uteri had not increased and remained comparable to noninfected controls. Four hours postinoculation, estrous rabbits.--At four hours postinoculation, the total number of fragmenting plasma cells (18, 19) and degranulating baSOphilic multilobulated cells (13) was slightly decreased (Plate 1, Fig. 3). The total numbers of mature plasma cells (16) remained fairly constant in contrast to the decreased numbers of eosino- philic plasma cells. Globular leucocytes (14, 15), which had increased markedly at one hour continued to decrease at two and again at four hours (Graph 1). The vasuclarity increased substantially for the first time at four hours. Accompanying this, was a moderate in— crease in peripheral blood leucocytes in the outer vessels. The total number of extravascular red blood cells continued to increase at four hours. It was at this time that blood polymorphonuclear leucocytes became evident as they migrated through the genital tract connective tissue. - Pseudopregnant uteri Control noninfected,_pseudgpregnant rabbits.--Pseudo- pregnant noninfected control uteri contained limited numbers of mature plasma cells (16) and chromatin remnants of termi- nal fragmenting cells (20). Other tissue cells, that is, 45 the basophilic preplasma cell (9, 10), basophilic lobulated preplasma cell (11), basophilic multilobulated cell (12), degranulating basophilic multilobulated cell (13), eosino— philic plasma cell, prefragmentation plasma cell (14), frag- menting plasma cell (18, 19) and globular leucocyte (14, 15) were not present in all pseudopregnant control uterine tis- sues examined (Table II). However, when tissues were stained with methylene blue, limited numbers of undifferentiated stem cells became evident. These cells were morphologically simi- lar to baSOphilic fibroblast-like cells (7, 8). Examination of pseudopregnant uterine tissues indicated that the chromatin remnants of terminal fragmenting cells (20) were confined to the epithelial and subepithelial portion of the uterus. Comparison of noninfected estrous uteri (Plate 1, Fig. 1) with noninfected pseudopregnant uteri (Plate 1, Fig. 4) illustrated the marked increase in uterine convulutions which characterized the pseudopregnant uteri. One hour postinoculation, pseudOpregnant rabbits.—-The general appearance of one hour pseudopregnant uteri was com- parable to control noninfected pseudopregnant tissues (Plate 1, Fig. 4). However, closer examination revealed that there was moderate increase in mature plasma cells (16). This was in contrast to the obvious paucity of eosinophilic and fragmenting plasma cells (l8, 19). The only additional cell types which were apparent in very low numbers were the baso— philic preplasma cells (9, 10) basophilic lobulated preplasma cells (12), and globular leucocytes (14, 15). However, while 46 these basophilic cell types were present, the total numbers were at an absolute minimum. Two hours postinoculation, pseudOpregnant rabbits.-- The general appearance of the two hour infected uteri ap— peared relatively unchanged from control and one hour in- fected uteri (Plate 1, Fig. 4). Mature plasma cells (16) continued to increase in number and appeared uniformly dis- persed throughout the endometrium. Accompanying this grad— ual increase was the realization of limited numbers of eosinophilic plasma cells which were apparent for the first time. There were still no prefragmenting (17) or fragment- ing plasma cells (18, 19) present in the two hour infected uteri. Staining with Alcian blue and modified Wright's stain demonstrated the presence of increasing numbers of baso- philic (9, 10) and basophilic lobulated preplasma cells (11). Globular leucocytes (14, 15) were present in numbers similar to those seen in the one hour stage of infection. The only cell type unique to this infection state was the basophilic fibroblast-like cell (7, 8) which was present in limited numbers and located directly beneath the uterine epithelium. The vascularity of the uterus at two hours appeared similar to one hour infected and control uteri. The addi- tion, no increase was observed in the numbers of blood leuco- cyles in outer blood vessels of the myometrium. 47 Four hour postinoculation, pseudgpregnant rabbits.-- Following an initial delay of two hours, the uterus of the four hour infected pseudopregnant rabbit was characterized by a pronounced increase of all cells of tissue origin as well as moderate numbers of blood leucocytes (Plate 1, Fig. 5). The substantial increase in mature (l6) and eosino- philic plasma cells accompanied the accumulation of moderate numbers of prefragmenting (17) and fragmenting plasma cells (18, 19) which were easily identified by their characteristic sausage and horseshoe shaped nuclei. Staining with hema— toxylin and eosin indicated that the fragmenting cells (18, 19) contained remnants of eosinophilic granules (Plate 1, Fig. 6). At the same time, there was a marked increase in baso- philic cell types. This was indicated by the pronounced ap- pearance of large numbers of baSOphilic multilobulated (l2) and degranulating basophilic multilobulated (13) cells. In fact, a large percentage of lobulated leucocytes in the stroma contained granules that were Alcian blue positive (pH 2.5-3.0) and burgundy in color with modified Wright's stain. Basophilic preplasma cells (9, 10) continued to in— crease at the same intensity noted at two hours (Graph 1). Basophilic fibroblast-like cells (7, 8) have increased slightly in the four hour infection when compared with the two hour infection. 48 Perhaps one of the most interesting alterations in response to bacterial antigen at the four hour stage was the moderate increase in globular leucocytes (14, 15) which was noted for the first time. It will be recalled that globular leucocytes increased in the estrous infected animal at one hour. The increase in globular leucocytes for the first time at four hours indicated a substantial delay in this modified morphogenesis in the pseudopregnant rabbit (Graph 1). The vascularity was noted to increase markedly at four hours but without any evidence of extravascular red blood cells. Accompanying the increase in vascularity was a simultaneous increase in polymorphonuclear leucocytes in the outer vessels of myometrium. In addition, blood leuco- cytes were observed in the endometrial tissues. Leucocytes of Tissue and Blood One difficulty in utilizing Alcian blue-hematoxylin staining is the inability to distinguish clearly between the degranulating baSOphilic multilobulated cells (13) and term- inal fragmentive cells (20) of tissue origin and the neutro- phils of peripheral blood. The reason for this is that some blood neutrOphils were found to contain minimum quantities of mucopolysaccharides which stain Alcian blue positive. This phenomenon was observed repeatedly whenever blood smears were fixed in Carnoy's solution and processed by the same procedures as genital tissues. As a result, a variety of 49 fixation and staining procedures were tested in an attempt to find a single stain which would clearly differentiate blood and tissue leucocytes. Of all the procedures utilized, one combination appeared promising. When tissues were fixed in Carnoy's solution (pH less than 4), autotechnicon pro— cessed, and subsequently stained with Wright's stain, the cell types previously described, that is, the basophilic fibroblast-like cell (7, 8), basophilic preplasma cell (9, 10), basophilic lobulated preplasma cell (11), and the baso- philic multilobulated cell (12), stained light burgundy. With this indication, a series of buffers were prepared and used in conjunction with alcohol-acetone dehydration follow— ing Wright's stain. Following ample modification, it was possible to distinguish these cells clearly from the sur- rounding tissues. The general tissue response (light pink) is decidedly different from the cells of tissue origin (burgundy). The cells of tissue origin could be identified under 43x microscopic power. Higher magnification clearly differentiates the plasma cell precursors, that is, the basophilic fibroblast-like cell (7, 8), basophilic preplasma cells (9, 10), basophilic lobulated preplasma cell (11), the basophilic multilobulated cell (13) from one another. Ad- ditional examination demonstrated the presence of identi- fiable amounts of burgundy granules in the mature plasma cell (16), eosinophilic plasma cell, prefragmentation plasma cell (17), fragmenting plasma cell (l8, l9) and terminal fragmenting cell (20). In essence, tissue leucocytes, as 50 well as their precursors, were found to contain granules which stain burgundy in color. The persistence of this burgundy color along with gradual morphological changes supported the concept of the presence of cell types having a continuity of development which culminates in death during fragmentation. As mentioned earlier, basal levels of degranulating basophilic multilobulated cells (13) were present in estrous rabbit noninfected uteri. These cells contained burgundy granules but in lesser numbers suggesting that the majority of granules had been released. On comparing the degranulat- ing basophilic multilobulated cells with blood polymorpho- nuclear leucocytes in the outer uterine vessels, it was found that the blood leucocytes were devoid of burgundy granules. More specifically, the absence of burgundy'granules in blood leucocytes of uterine vessels and the presence of them in tissue leucocytes indicates that blood and tissue leucocytes can be made cytochemically distinct. 51 .h>mmn ham> .++++ mh>mmn +++ mopmnmpoe x++ mumpdefia H mucmpfi>mcoc .I "msamps on» CH pmpazooo pm: fleas; mpfi>fiuom Hapaoaaopomn mo mama map mm Ham: mm soESH can Esfinpmaopcm one oucfi soaumnmaa oaphooosma no mopwmo on» pcommpama I use + maonezm one +++ ++ ++++ : I I + m mpa>upom I I I H HmUAOHAmpomm +++ ++ ++++ : + + + m I I I a mmpmooosma I I I o Hmcaesq ++ +++ ++++ : + + ++ m + I + H mHmOpmooosmH I I H o Hafinpmsopsm pmuanuomHnm>o pcmcwmnaoosomm mBOApmm mEHB EmpH coaumHSoocH pannmm no mass .mpfi>fipom Hmnfiofinmpomn madam»: m>fipmamQEoo Sufi: “coapommcfi Hmfiampomn wcHSOHHom pan on poand xswag was Esanpmaopcm scans»: on» owed coaumswas mpzooosma wsH>Ho>ca .mpfioomn pmNHEOpooHam>o cam .pcmcwmnoopsmmd .msoapmm macaw comfiamQEoo ¢II.H mqm¢e TABLE 2.-Effects of inoculation of bacterial antigen on cell types present in uterine genital tissues. fiqtaexnoseA ensstq eAtqoeuuoo ut squoooneI pootg SIQSSGA Jeqno u: squooonet pootg (08) II90 in: -quemSeJJ Ieutmaeg (6t ‘8I) Iteo emsetd SUIQUQwSBJfl (LT) 1130 emsstd uotqequemSeJJeJa Itao emsetd OIIIudOUIsos (9t) 1190 ewsetd eanqsw (SI ‘fiI) eqfiooonet aetnqoto (ET) II90 99491 -HQOIIQInm OTIqu -oseq SutqetnueJSeq (BI) II90 paaeIHQOI ~14Inm atttqdoses (II) II90 esmetdaad page: -nq0I attrqdosea (0T ‘6) 1190 amsetd -ead otttqdoseg (8 ‘1) II90 SHIT-ASEIQ -OJQIJ Otttqdoses Infection Time 52 Estrous Control One hour Two hour Four hour Pseudopregnant Control One hour Two hour Four hour not evident; 1 = limited; 2 = slight; 3 = moderate; 4 = heavy; 0 KEY: 5 = very heavy. 53 .QOHpommcH HMthp0Mh poodpcfi mafiazp can on poand Hump: panama pcmcwmpdopzwmd cam wBOABmm CH mmpzoooswa hmadnoaw cam maamo mammammna ofiaficdommmII.H mm