Innervation and gap junction formation in the myometrium of pregnant little brown bats Myotis lucifugus.код для вставкиСкачать
THE ANATOMICAL RECORD 221~611-618(1988) Innervation and Gap Junction Formation in the Myometrium of Pregnant Little Brown Bats, Myotis lucifugus G.D. BUCHANAN, R.E. GARFIELD, AND E.V. YOUNGLAI School of Nursing (G.D.B.)and Departments of Neurosciences (R.E.G.) and Obstetrics and Gynecology (R.E. G., E. V Y), Faculty of Health Sciences, McMaster University, Hamilton, Ontario, Canada L8N 325 ABSTRACT Pregnant Myotis lucifugus were captured in mist nets set outside a large maternity colony and, in most cases, were examined 12-15 hours later. Anterior and posterior halves of uteri were pinned to dental wax and either incubated in glyoxylic acid to produce adrenergic nerve fluorescence, or fixed in buffered glutaraldehyde for electron microscopy. Blood was collected for radioimmunoassay of plasma progesterone. We found no evidence of decreased nerve fluorescence even in late pregnancy when plasma progesterone levels were 10-20 times preovulatory values. Ultrastructural examination also showed no evidence of damage to, or destruction of, nerves in the myometrium. However, we did find gap junctions between myometrial muscle cells during the periparturient period in both normal and aborting bats. Gap junctions began to form several days before term, increased in number until parturition, then decreased dramatically within a day or two thereafter. The percentage of muscle cell plasma membrane involved in gap junction formation was closely correlated with plasma progesterone levels, although whether this is causal or coincidental is not certain. These data do not agree with the conclusions drawn from observations in other mammals that a disappearance of adrenergic nerves from the myometrium is associated with the initiation of parturition, or that gap junction formation is associated with changes in nerve function. They do, however, lend further support to the hypothesis that there is neurogenic control of myometrial contractility in M. lucifugus uteri at term. A variety of roles have been ascribed to myometrial contractility during the reproductive cycle, including sperm transport (Austin, 1974) and positioning of the conceptus at the implantation site (Boving, 1972), and expulsion of the uterine contents at parturition (Huszar and Roberts, 1982). Whereas myometrial function during parturition has been much studied, the factors governing this extraordinarily well organized process are incompletely understood (Liggins, 1979; Csapo, 1981). Hypophyseal hormones, ovarian hormones, and prostaglandins modulate myometrial activity (Fuchs, 19781, but earlier studies indicated that neurogenic modulation was either absent or minimal (Reynolds, 1965).Certainly central nervous control is not essential, since neither uterine denervation nor spinal cord section prevents parturition (Bell, 1972). However, these procedures do not interrupt postganglionic neurons originating in the paracervical ganglion of Frankenhauser (Thorbert, 1978). Since these and other neurons are affected by steroid hormones, it is possible that they play some intermediary role in the control of myometrial contractility (Owman, 1981).There is also evidence that autonomic nerves to the female reproductive tract affect cyclic variation in gonadotropins and steroid hormones (Burden et al., 1981). 0 1988 ALAN R. LISS, INC In several mammalin species, the uterus is innervated primarily by postganglionic, adrenergic neurons that are in close association with blood vessels and myometrial muscle cells (Marshall, 1981; Owman et al., 1986). Cholinergic innervation is relatively sparse and also associated with blood vessels and myometrial cells (Papka et al., 1985; Garfield, 1986).Extensive studies in the guinea pig (see Owman, 1981) by ultrastructural, histofluorescent, and biochemical methods show that adrenergic nerves disappear from the uterus during late pregnancy. In reviewing these, as well as studies from other species, Marshall (1981) concluded that the data collectively suggested a local degeneration of adrenergic axons in the uterus at term. Whereas loss of adrenergic nerves may explain in part increased myometrial contractility at term, it does not explain the high degree of coordinated activity during parturition. There is increasing evidence, however, that gap junctions play a significant role in myometrial coordination by converting the myometrium into a functional syncytium. Gap Received February 27,1987; accepted November 24,1987. Address reprint requests to Dr. G. Dale Buchanan, Faculty of Health Room 2532, HamSciences, McMaster University, 1200 Main St. W., ilton, Ontario, Canada L8N 325. 612 G.D. BUCHANAN ET AL. junctions are rarely found among myometrial smooth muscle cells except in the periparturient period, whether this be at term or premature (Garfield, 1985). Thus, there is an inverse relationship between disappearance of adrenergic nerve fibers and appearance of gap junctions in the myometrium. However, whether these events are causally related or simply represent correlated responses to some other, possibly hormonal change is not entirely clear. To our knowledge, myometrial ultrastructure has not been examined in bats except for our earlier study of hibernating, nonpregnant Myotis lucifugus, the common little brown bat (Buchanan and Garfield, 1984).We found the myometrium to be densely supplied by adrenergic nerve fibers with axonal varicosities closely associated with smooth muscle cells as well as blood vessels. Nonadrenergic nerve fibers were mostly associated with blood vessels; however, 10-15% of axonal varicosities found in close relation to myometrial muscle contained small agranular (presumptive cholinergic)secretory vesicles. Injection of estradiol, progesterone, or both did not alter the appearance of adrenergic fibers or elicit gap junction formation. Injection of 6-hydroxydopamine, which depletes catecholamine stores and damages adrenergic nerves (Kostrzewa and Jacobowitz, 1974), did produce evidence of nerve damage but did not elicit gap junction formation. Based on these results, we speculated that neurogenic control of the myometrium might be relatively more important in bats than in the other species that have been studied. In the present study, we examined myometria of M. lucifugus during pregnancy, at parturition, and while undergoing abortion. The results show no evidence of neuronal loss or degeneration in late pregnancy even when gap junctions are present, and further support the hypothesis of a neurogenic modulation of the myometrium of M. Zucifugus. MATERIALS AND METHODS Animals and Tissues Pregnant Myotis lucifugus lucifugus were captured in mist nets set outside a large maternity roost at Red Bay, Ontario (lat. 44"48'N), brought to Hamilton in small wooden cages (supplied with water), and usually examined the next morning. Bats kept for more than 24 hours were maintained in a warm (28-30°C) room and fed mealworms (Tenebrio molitor larvae) daily. Before autopsy, bats were weighed and then sacrificed by cervical fracture. The thoracic cavity was opened quickly and blood was collected from the pectoral and subclavian veins in heparinized capillary tubes. Reproductivetracts and adnexa were excised and photographed, after which uteri were divided into dorsal and ventral halves and processed as described previously (Buchanan and Garfield, 1984). Briefly, tissues were spread and pinned to dental wax serosal side down and then incubated as outlined below. Equal numbers of anterior and posterior halves were allocated to each procedure. Histofluorescence Spread tissues were incubated for 30 minutes at 4°C in a solution of 2.0 gm glyoxylic acid, 0.92 gm HEPES buffer, and 37 gm sucrose per 100 ml at pH 7.4. Glyoxylic acid reacts with catecholamines and, in the absence of water, produces a bright fluorophore considered specific for adrenergic nerves (Llewellyn-Smithet al., 1981; Tsu- kahara et al., 1982). Incubated tissues were partly dried in a cool air stream, transferred to glass slides, and dried for 1-2 hours. Dried tissues were incubated at 100°Cfor 4 minutes, mounted in mineral oil under glass coverslips, and examined on a Zeiss Large Universal fluorescent microscope equipped with an epifluorescent condenser (IIIRS),a BP43615 excitation filter, an LP473 barrier filter, and a 460 beam splitter. To estimate nerve density, photomicrographs of fluorescent nerves were printed at a magnification of 495 x such that each photograph represented 0.17 mm2 of uterine wall. A grid containing 84 lines, 2 cm long, and 2 cm apart on the transverse and vertical axes was placed over the photograph and the number of intersects of fluorescent fibers with grid lines counted. Total uterine surface area was estimated by measuring the long and short dimensions of the uterus from photographs taken at autopsy and substituting one-half their mean value (r)in the formula 47r? (surface of a sphere). Electron Microscopy To minimize distortion and intersample variation due to preparatory procedures, tissues were incubated in Kreb's buffer (pH 7.4) containing 0.05 mM pargyline for 1hour at 37°C before fixation. Pargyline (Sigma Chemicals, St. Louis, MO), a monoamine oxidase inhibitor, prevents breakdown of norepinephrine during incubation. Incubated tissues were fixed in cacodylate-buffered glutaraldehyde, postfixed in osmium tetroxide, stained en bloc with many1 acetate, and embedded in Spurrresin. Sections were cut with a diamond knife, mounted on 200-mesh grids, stained with lead acetate, and examined in a Philips 301 transmission electron microscope. Gap junctions were quantitated by the method of Puri and Garfield (1982). Plasma Progesterone Blood samples were spun for 10 minutes in a microhematocrit centrifuge and plasma samples were frozen at -20°C until assayed. Progesterone content of individual plasma samples was determined by radioimmunoassay using an antibody produced in rabbits against progesterone-11-a-hemisuccinyl-bovine serum albumin, and (l,23H) progesterone as the labeled ligand. Details of the assay procedure are described elsewhere (Buchanan and Younglai, 1986). RESULTS Uterine Appearance Like other holarctic vespertilionid bats, Myotis Zucifugus completes folliculogenesis and mates before entering hibernation, but ovulation and fertilization are postponed until 1-3 days after arousal the next spring. Pregnancy lasts approximately 8 weeks and can be divided into 5 stages recognizable by uterine shape and width (Buchanan and Younglai, 1986). Whereas the lengths of Stages I and I1 have been established (Buchanan, 1987), the time period given below for later stages are approximations derived from periodic field collections. In Stage I (preimplantation, days 1-10], uteri may show slight hyperemia and edema, but do not differ from preovulatory uteri in shape or width (4 mm). In Stage I1 (early implantation, days 10-14), the right uterine horn becomes highly domed antimesometrially. During Stage I11 (embryogenesis, weeks 3 and 4), uteri GAP JUNCTIONS AND NERVES IN PREGNANT BAT MYOMETRIA increase from 4 to 7 mm wide and the right uterine horn becomes ovoid, but the left horn is still distinct. Uterine width increases to 12 mm during Stage IV (early fetal period, weeks 5 and 6) and the entire uterus becomes smoothly ovoid as the left horn is incorporated into the uterine wall (Fig. 1). Growth is quite rapid in Stage V (late fetal period, weeks 7 and 8)and the thinly stretched uterine wall begins to conform to the fetal contours (Fig. 2). Uterine width will reach 21-23 mm at term. Fluorescent Microscopy Examination of uterine preparations incubated in gyloxylic acid showed an extensive retiform pattern of adrenergic nerve fibers with numerous axonal varicosities. There were no differences in either nerve density or fluorescence between mesometrial and antimesometrial or dorsal and ventral regions. At midpregnancy (Fig. 31, nerve density-the number of nerve fibers per unit area-did not appear to differ from nerve density in earlier specimens. However, in uteri near term (Fig. 41, nerve density was somewhat reduced, although fluorescence of individual fibers was not detectably changed. Computation of nerve fiber intersects with a standard grid (Table 1)confirm that nerve density was, indeed, statistically less in term uteri. However, the increase in uterine surface area during pregnancy more than offsets this decrease in density. In fact, computation of nerve fiber intersects per entire uterus yields the implausible result that there were severalfold more nerve fibers at term than during early pregnancy. Electron Microscopy Numerous unmyelinated axons were found in close association with smooth muscle cells of the myometrium during pregnancy (Fig. 5) and at term (Fig. 6). The majority of the axons bore varicosities containing small (c. 50 nm diam) dense-cored neurosecretory vesicles (Fig. 7), which are considered indicative of adrenergic nerves. When pregnant bats (early Stage IV)were treated with 5-hydroxydopamine (0.8 mgldayl3 days), many of the granular vesicles exhibited sharply increased electron density and some were enlarged (Fig. 8). Most of the remaining axonal varicosities contained small (c. 50 nm diam) agranular vesicles, which are characteristic of cholinergic nerves. We also found axonal varicosities containing large opaque or large dense-cored neurosecretory vesicles (Fig. 9). The biochemical function of these latter nerves is undetermined, although they resemble the purinergic vesicles that have been described by Burnstock (1980). Gap junctions (Figs. 6,lO) were present in the myometria of normal bats examined during the last quarter of pregnancy, but were not seen earlier except in two bats examined during Stage 111. Since gap junctions were found also in the myometria of bats undergoing abortion during Stages 111, IV,and V, we suspect that these bats were about to abort. However, their plasma progesterone levels were within the normal range, and we found no morphological evidence (ischemia, intrauterine bleeding, wrinkling of the uterine wall) to justify excluding them. Figure 11compares plasma progesterone levels and gap junction area, expressed as a percent of plasma membrane area (Puri and Garfield, 1982), in bats from various stages of pregnancy. As shown, gap junctions were present in uteri 11 mm wide and larger, 613 increased in area until term, and declined rapidly postpartum. Figure 11 also shows that the change in gap junction area paralleled the change in plasma progesterone levels. DISCUSSION Most knowledge of mammalian uterine innervation is derived from studies of guinea pigs, rabbits, dogs, cats, and humans in which myometrial innervation is primarily, if not exclusively, adrenergic, and in all of which there is a progressive loss of demonstrable nerve fibers during pregnancy. It has been hypothesized that adrenergic nerves play a role in maintaining uterine quiescence during pregnancy by inhibiting contractility, and that their loss permits, but does not necessarily initiate, increased contractility at term (Thorbert, 1978; Marshall, 1981). However, this hypothesis is difficult to reconcile with recent studies showing that rat myometrium is innervated principally by cholinergic fibers and a lesser number of adrenergic and peptidergic fibers (Papka et al., 1985) or nonadrenergic, noncholinergic fibers (Garfield, 1986).Moreover, the theory that the loss of nerves contributes to events leading to parturition is clearly incompatible with the results of the present study, which showed no evidence of disappearance of uterine nerves in pregnant M. lucifugus even at term when gap junctions are present. As noted in Table 1,fluorescent nerve fiber density in near-term uteri was only 60% of density in early pregnancy. However, due to the increase in uterine surface area, the total number of nerve fibers per uterus appears to be increased considerably. In fact, the calculated number of adrenergic fibers in early and midpregnancy is probably greatly underestimated since the uterine wall is much thicker than in late pregnancy and many fibers are either above or below the focal plane of the photographs. In any case, there was unquestionably no general loss or disappearance of adrenergic fibers as has been described in other species. This conclusion is also supported by electron microscopic examination of pregnant uteri, which showed an abundance of unmyelinated nerve fibers closely associated with muscle cells even when gap junctions have formed (Fig. 6). Most axonal varicosities contained small dense-cored secretory vesicles (Figs. 5, 71, which are conceded to contain catecholamines and, hence, be indicative of adrenergic fibers (Burnstock, 1980).Moreover, after treatment with 5-hydroxydopamine,which accumulates in the secretory vesicles of adrenergic neurons by acting as a false transmitter (Tranzer and Thoenen, 19671, axonal varicosities containing very dense, occasionally enlarged vesicles were seen (Fig. 8). Injections of 6-hydroxydopamine, which damages and destroys adrenergic nerves (Kostrzewa and Jacobowitz, 1974; Buchanan and Garfield, 1984), were unsuccessful as even minute doses were highly toxic t o pregnant bats. The presence of gap junctions between myometrial muscle cells during the latter part of pregnancy (uteri 11mm wide and larger), and their disappearance shortly after parturition, agrees generally with observations in other species, including humans (reviewed by Verhoeff and Garfield, 1986). However, the implication in Figure 11 that this process begins 2 weeks before term in M. lucifugus must be viewed with reservation since in other species gap junctions appear only 1to 2 days prepartum 614 G.D. BUCHANAN ET AL. Fig. 1. Ventral aspect of a uterus of M. lucifugus during the early fetal stage (Stage IV,uterine width 9 mm). The greatly enlarged right uterine horn and uterine body are smoothly ovoid and the left uterine horn (small arrow) is partially incorporated into the uterine wall. The cervix is indicated by the large arrow. x4.8. Fig. 2. Anteroventral view of a uterus of M. Zucifugus approaching term (Stage V, uterine width 16 mm). A ridge containing enlarged blood vessels (small arrows lying along the antimesometrial margin bisects the placental area). The position of the cervix (not visible) is indicated by the large arrow. The fetal head lies to the left in the photograph. ~ 4 . 8 . Fig. 5. A low magnification electron micrograph of the myometrium of a bat at midpregnancy (Stage 111, uterine width 7 mm). A nerve lying in close proximity to the muscle cells has several varicosities (arrows) that contain small densecored secretory vesicles characteristic of adrenergic nerves. x 18,000. Fig. 3. Fluorescent micrograph of a uterine wholemount from a bat during the early fetal stage (Stage IV, uterine width 9 mm) after incubation with glyoxylic acid. The tissue contains an abundant network of fluorescent (adrenergic) nerve fibers with numerous varicosities (arrows). The fibers in the upper right, which appear dim, are below the focal plane of the photograph. X 160. Fig. 4. Fluorescent micrograph of a uterine wholemount from a bat near term (Stage V, uterine width 21 mm) after glyoxylic acid incubation. Numerous fluorescent (adrenergic) fibers bearing varicosities are present (arrows). Although fluorescence of individual fibers is not decreased, fewer fibers are visible in the picture (compare to Fig. 3) probably due to stretching of the uterine wall. X 160. Fig. 6. A low magnification electron micrograph of the myometrium of a bat near term (Stage V, uterine width 20.5 mm). Small nerves containing morphologically normal axons (small arrows) occur in close approximation to smooth muscle cells. Note gap junction (large arrow) between two muscle cells. x 18,000. GAP JUNCTIONS AND NERVES IN PREGNANT BAT MYOMETRIA 615 616 G.D. BUCHANAN ET AL. (Puri and Garfield, 1982; Garfield, 1985). Because pregnancy is asynchronous by 3 to 4 weeks within a colony of M. lucifugus (Tuttle and Stevenson, 1982), Buchanan and Younglai (1986) estimated the lengths of Stages IV and V by plotting the midpoint of the interval during which specific stages were collected. The precision of 3 806 f 8b 16.0 f 2.1' Early 4 701 k 107b 96.0 14.gd this method is limited especially in late pregnancy when Mid 5 489 f 46" 940.6 k 113.1e growth is very rapid, and we suspect, therefore, that Late Stage V (uterine width 13 to 22+ mm) is actually much aTechniques for calculating intersects and for computing surface area shorter than the 2 weeks depicted in Figure 11. are described in Materials and Methods. Gap junctions may serve to synchronize and propagate b,c,d,eValues in same column without common superscripts differ statistically by Duncan's Multiple Range Test (p < 0.05). myometrial contractions during labor, or to enhance TABLE 1. Fluorescent nerve fiber density per unit area in uteri of pregnant Myotis tueifugusa Stage of No. X intersects X surface pregnancy bats per mm2 area (mm? Fig. 7. Electron micrograph showing two nerve varicosities (arrows) Fig. 9. Electron micrograph of a nerve varicosity from the same containing small dense-cored secretory vesicles that are characteristic uterus shown in Figure 7. The varicosity contains large dense-cored or of adrenergic nerves. The varicosities are in close proximity to a mus- large opaque neurosecretory vesicles. The function of these vesicles is cle cell (right of picture). ~ 3 0 , 0 0 0 . undetermined, but they resemble the purinergic vesicles described by Burnstock (1980). ~ 3 0 , 0 0 0 . Fig. 8 . Nerve varicosity in the myometrium of a pregnant M. Zuczfugus (Stage IV,uterine width 8 mm) treated with 5-hydroxydopamine Fig. 10. High magnification electron micrograph showing a gap for 3 days. Note increased density and enlargement of neurosecretory junction formed between two smooth muscle cells in the myometrium vesicles (compare to Fig. 7). x 30,000. of a bat near term. The limits of the gap junction are indicated by arrows. x 100.000. 617 GAP JUNCTIONS AND NERVES IN PREGNANT BAT MYOMETRIA 11 Fig. 11. Myometrial gap junction area and plasma progesterone nant, hibernating bats examined in a n earlier study (Buchanan and levels in pregnant and postpartum M. lucifugus. Specimens were Garfield, 1984)are included. Solid circles indicate gap junctional area, grouped by pregnancy stage and uterine width as described by Buch- as a percent of plasma membrane area, for individual bats. Gray bars anan and Younglai (1986).Postpartum bats had blood in their vaginae, indicate mean plasma progesterone values for all bats in each size indicating that parturition had taken place the day of capture. Lactat- group. The 2 Stage I11 bats (5 mm and 5.5 mm uterine widths) that ing bats had delivered 3 or more days before examination, as judged had gap junctions may have been about to abort (see Results). by degree of uterine involution. For comparison, data from 4 nonpreg- myometrial response to chemical mediators (Garfield, other chiropterans is unknown; however, several factors 1985). Whatever their precise role, it is likely that the suggest that precise control of myometrial contractility presence of gap junctions is regulated by changes in may be of particular value to heterothermic bats. First, steroid hormone or prostaglandin levels (MacKenzie and due to their high energy requirements, especially during Garfield, 1985) since changes in these hormones accom- late pregnancy and lactation (Studier et al., 1973), M. pany and are generally conceded to govern the onset of lucifugus can ill afford t o forego daily foraging; and both normal and premature labor (Thorburn and Chal- evidence from a closely related species, Pipistrellus p i p lis, 1979). Indeed, in vivo and in vitro studies in the rat istrellus, suggests that if they do so they revert to torpor show that progesterone inhibits, whereas estrogen stim- (Racey, 1974). Second, although pregnant M. lucifugus ulates gap junction formation (Garfield et al., 1980, remain away from the maternity (day) roost the entire 1982).In rats there is a prepartum decline in circulating night, occupying separate night roosts in the interim progesterone and a rise in estrogens, so it is uncertain if between feeding bouts (Anthony et al., 19811, there is no either progesterone withdrawal per se or the resultant evidence that parturition occurs elsewhere than in the increase in estrogedprogesterone level triggers gap maternity roost. It is logical to infer, therefore, that junction formation (Puri and Garfield, 1982).In humans, parturition in M. lucifugus is subject to rigid temporal plasma progesterone does not decline before gap junc- control, which may be achieved by myometrial nerves tion appearance (Garfield and Hayashi, 1981);however, acting to either inhibit or stimulate contractility when estrogen levels rise Vhorburn and Challis, 1979). M. appropriate. Zucifugus resembles humans in that gap junctions form while progesterone levels are still elevated (Fig. 11). ACKNOWLEDGMENTS Unfortunately, plasma estrogen values are not available, so the possibility that the ratio of estrogen to pro- We thank Francesca C. Bullock for expert technical help with autopsies and preparation of tissues for fluresgesterone is altered is undetermined. In conclusion, Myotis lucifugus differs from other cent microscopy, and Debbie Merrett for preparation mammals thus far studied in that myometrial nerves do and examination of tissues by electron microscopy and not disappear during pregnancy even in the peripartu- preparation of photographs. We especially thank Mr. rient period when gap junctions appear. This is concor- Harve Whitcroft, owner of Red Bay Outfitters, and his dant with our previous suggestion (Buchanan and family for permitting us to collect bats on their premises Garfield, 1984)that in this bat the myometrium may be and for their kind hospitality and interest in our resubject to greater neurogenic control than in other mam- search. This work was supported by grants from the mals. Whether myometrial innervation is similar in Medical Research Council of Canada. 618 G.D. BUCHANAN ET AL. LITERATURE CITED Anthony, E.L.P., M.H. Stack, and T.H. Kunz 1981 Night roosting and nocturnal time budget of the little brown bat, Myotis lucifugus: Effects of reproductive status, prey density and environmental conditions. Oecologia (Berl.), 51:151-156. Austin, C.R. (1974)Recent progress in the study of eggs and spermatozoa: Insemination and ovulation to implantation. In: Reproductive Physiology, R.O. Greep, ed. MTP International Review of Science, Physiology, series 1,vol. 8,Butterworth‘s, London, p. 96. Bell, C. 1972 Autonomic nervous control of reproduction: Circulatory and other factors. Pharmacol. Rev., 24:657-736. Boving, B.G. 1972 Spacing and orientation of blastocysts in utero. In: Biology of Mammalian Fertilization and Implantation. K.S. Moghissi and E.S.E. Hafez, eds. C.C. Thomas, Springfield, IL, pp. 357- 378. Buchanan, G.D. 1987 The timing of ovulation and early embryonic development in Myotis l u c i g u s (Chiroptera:Vespertilionidae) from northern central Ontario. Am. J. Anat., 178:335-340. Buchanan, G.D., and R.E. Garfield 1984 Myometrial ultrastructure and innervation in Myotis lucifugus, the little brown bat. Anat. Rec., 210:463-475. Buchanan, G.D., and E.V. Younglai 1986 Plasma progesterone levels during uremancy in the little brown bat Myotis lucifugus (Vespertilionrddae).-Biol. -%prod., 34:878-884. Burden, H.W., I.E. Lawrence, T.M. Louis, and C.A. Hodson 1981 Effects of abdominal vagotomy on the estrous cycle of the rat and the induction of pseudopregnancy. Neuroendocrinology, 33:218-222. Burnstock, G. 1980 Cholinergic and purinergic regulation of blood vessels. In: Handbook of Physiology, Section 2:The Cardiovascular System, Vol. 11, Vascular Smooth Muscle. D.F. Bohr, A.P. Somlyo, and H.V. Sparks, Jr., eds. Am. Physiol. Soc., Bethesda, pp. 567- 612. Csapo, A.I. 1981 Force of labor. In: Principles and Practice of Obstetrics and Perinatology, L. Iffy and H.A. Kiminetzky, eds. John Wiley and Sons,New York, pp. 761-799. Fuchs, A.R. (1978)Hormonal control of myometrial function during pregnancy and parturition. Acta Endocrinol. (Suppl.) (Copenhagen), 221:l-70. Garfield, R.E. 1985 Cell-tocell communication in smooth muscle. In: Calcium and Contractility. A.K. Grover and E.E. Daniel, eds. Humana Press, Clifton, NJ, pp. 143-173. Garfield, R.E. 1986 Structural studies of innervation in nonpregnant rat uterus. Am.J. Physiol., 251:C41-C54. Garfield, R.E., and R.H. Hayashi 1981 Appearance of gap junctions in the myometrium of women during labor. Am. J. Obstet. Gynec., 140:254-260. Kostrzewa, R.M., and D.M. Jacobowitz 1974 Pharmacological actions of 6-hydroxydopamine.Pharmacol. Rev., 26:199-288. Liggins, G.C. 1979 Initiation of parturition. Br. Med. Bull., 35145- 150. Llewellyn-Smith, I.J., A.J. Wilson, J.B. Furness, M. Costa, and R.A. Rush 1981 Ultrastructural identification of noradrenergic axons and their distribution within the enteric plexuses of the guinea-pig small intestine. J. Neurocytol., 10:331-352. MacKenzie, L.W., and R.E. Garfield 1985 Hormonal control of gap junctions in myometrium. Am. J. Physiol., 248:C296-C308. Marshall, J.M. 1981 Effects of ovarian steroids and pregnancy on adrenergic nerves of uterus and oviduct. Am. J. Physiol., 24032165- C174. Owman, C. 1981 Pregnancy induces degenerative and regenerative changes in the autonomic innervation of the female reproductive tract. In: Development of the Autonomic Nervous System, K. Elliot and G. Lawrenson, eds. Ciba Found. Symp., 83:252-279. Owman, C., P. Alm, N.-0. Sjoberg, and J. Stjernquist 1986 Structural, biochemical, and pharmacological aspects of the uterine autonomic innervation and its remodeling during pregnancy. In: The Physiology and Biochemistry of the Uterus in Pregnancy and Labor. G. Huszar, ed., CRC Press, Boca Raton, FL, pp. 5-23. Papka, R.E., J.P. Cotton, and H.H. Traurig 1985 Comparative distribution of neuropeptide tyrosine-, vasoactive intestinal polypeptide, substance P-immunoreactive, acetylcholinesterase-positiveand noradrenergic nerves in the reproductive tract of the female rat. Cell Tissue Res., 242475-490. Puri, C.P., and R.E. Garfield 1982 Changes in hormone levels and gap junctions in the rat uterus during pregnancy and parturition. Biol. Reprod., 27:967-975. Racey, P.A. 1973 Environmental factors affecting the length of gestation in heterothermic bats. J. Reprod. Fertil., Suppl. 19:175-189. Reynolds, S.R.M. 1965 Physiology of the Uterus, 2nd ed. Hafner, New York. Studier, R.H., V.L. Lysengen, and M.J. O’Farrell 1973 Biology of Myotis thysunodes and M. lucifigus (Chiroptera: Vespertilionidae). 11. Bioenergetics of pregnancy and lactation. Comp. Biochem. Physiol., 444:467-471. Thorbert, G. 1978Regional changes in structure and function of adrenergic nerves in the guinea-pig uterus during pregnancy. Acta Obstet. Gynecol. Scand. (Suppl.), 79:l-32. Thorburn, G.D., and J.R.G. Challis 1979 Endocrine control of parturition. Physiol. Rev., 59:863-918. Tranzer, J.P., and H. Thoenen 1967 Electronmicroscopic localization of 5-hydroxydopamine (3,4,5,-trihydroxyphenylethylamine)a new “false” transmitter. Experientia, 2:743-745. Tsukahara, S., S. Kobayashi, K. Sugita, and T. Nagata 1982 Histochemical study of the postnatal development of autonomic nerves in the mouse iris, using a whole-mount preparation method. Histochemistry, 74:481-486. Tuttle, M.D., and D. Stevenson 1982 Growth and survival of bats. In: Ecology of Bats. T.H. Kunz, ed. Plenum Press, New York, pp. 105- Garfield, R.E., M.S. Kannan, and E.E. Daniel 1980 Gap junction formation in myometrium. Control by estrogens, progesterone, and prostaglandins. Am. J. Physiol., 238:C81-C89. Garfield, R.E., C.P. Puri, and A.I. Csapo 1982 Endocrine, structural and functional changes in the uterus during premature labor. Am. 150. Verhoeff, A., and R.E. Garfield 1986 Ultrastructure of the myometrium J. Obstet. Gynecol., 142:21-27. and the role of gap junctions in myometrial function. In: The Huszar, G., and J.M. Roberts 1982 Biochemistry and pharmacology of Physiology and Biochemistry of the Uterus in Pregnancy and Lathe myometrium and labor: Regulation at the cellular and molecbor. G. Huszar, ed. CRC Press, Boca Raton FL, pp. 73-91. ular levels. Am. J. Obstet. Gynec., 142:225-237.