|
|
||||||||
Articles |
a Institute of Nutritional Science, University of Potsdam, D-14558 Bergholz-Rehbrücke, Germany
b Department of Physiology, Veterinary Faculty, University of Leipzig, D-04103 Leipzig, Germany
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
Specific fluorescence in the endometrium of pigs was described during early gestation from Day 12 to Day 40, but not in cyclic pigs between Days 15 and 19 of the estrous cycle [5]. The authors described two different types of fluorescence under UV exposure, with a peak emission at excitation of 365 nm associated with the mucosa of the opened uterus. Both the yellowish-greenish and the reddish fluorescence showed changes in intensity and location over time. Until now, neither of the observed patterns of fluorescence has been characterized. On the basis of the description of color and the UV excitation wavelength used, this observed autofluorescence may be attributable to vitamin A. This characteristic of vitamin A was used many years ago to describe the distribution of vitamin A in different tissues [6, 7]. The observation of vitamin A in the porcine uterus could be explained functionally because both vitamin A, as retinol, and its specific carrier protein, the RBP, are major components of the uterine histotroph that supplies the developing conceptus with vitamin A for morphogenesis [8].
The aim of the present study was therefore to evaluate the known autofluorescence [5] in uterine tissue with regard to its possible association with vitamin A. Furthermore, the study was conducted to characterize the changes in the content and distribution of vitamin A and the distribution of proteins specific in vitamin A metabolism, such as the cellular retinoic acid-binding protein (CRABP) I and the retinoid X receptor ß (RXRß), on the cellular level in the uterus of pigs during early gestation.
| MATERIALS AND METHODS |
|---|
|
|
|---|
The uteri of a total of 30 crossbred primiparous gilts were removed after slaughter between Day 14 and Day 30 of gestation at 2-day intervals (34 animals per group). Conception rate of the gilts was 76.7%. Uteri were cut open and fluorescence was observed macroscopically under UV exposure (excitation wavelength: 312 nm). The emission spectrum was measured for 5 sec using a high-sensitivity (5.4 x 1014 counts/watt second) CCD (charged coupled device) spectral module (M. 1000 JETI GmbH Jena; Fa. H. Pannhoff, Optische Messtechnik, Hamburg, Germany). For the quantification of vitamin A in uterine tissue by HPLC, sections of implantation sites and nonimplantation sites, as well as specific samples of high-fluorescence areas, were obtained. Prior to sampling, embryos were removed mechanically or by flushing. Samples were separated prior to freezing into endometrium and myometrium. All samples were kept frozen at -80°C until analyzed.
Fluorescence Microscopy
For fluorescence microscopy, sections (6 µm) were prepared from native frozen tissues with a freezing microtome at -20°C. The unfixed and unstained specimen was observed with a reflected-light fluorescence microscope (Model BX40; Olympus, Tokyo, Japan). A 330- to 385-nm exciter filter and a 420-nm barrier filter were used (Cube U-MWU, Olympus). Frame and focus were controlled under visible light. Autofluorescence was documented by photography (Fujichrome Provia, ASA 1600; Tokyo, Japan). The same section was stained with hematoxylin-eosin (HE) and photographed (Fujicolor Super G, ASA 200). The association of the fluorescence with lipids was tested by treatment of cryosections with n-hexane [9].
Specificity of Antisera
Normal rat serum, monoclonal (mouse) anti-RXRß antibody (IgG1, MOK 1317, 2 mg/ml in PBS), monoclonal (mouse) anti-CRABP I antibody (IgG2b, C-1), affinity pure rat anti-mouse IgG (H+L; 1.2 mg/ml), and mouse peroxidase-anti-peroxidase (20 mg/ml) were purchased from Dianova GmbH (Hamburg, Germany); normal swine serum, rabbit anti-human RBP (8.2 g/L), swine anti-rabbit serum (160 g/L), and rabbit peroxidase-anti-peroxidase were purchased from Dako Diagnostica (Hamburg, Germany). All antisera were used as recommended by the company.
Immunohistochemistry
The immunolocalization of RBP, CRABP I, and RXRß was performed in uterine samples collected immediately after slaughter and then placed in 4% formaldehyde solution. Samples were dehydrated through graded series (5096%) of ethanol, embedded in paraffin, sectioned at 24 µm, and mounted on glass slides (Super-Frost; Fa. Menzel, Braunschweig, Germany). RBP, CRABP I, and RXRß were localized by using the peroxidase-anti-peroxidase method (PAP) as described previously [10]. Slides were deparaffinized, rehydrated in ethanol, and incubated for 45 min in 0.5% H2O2 in 100% methanol to inactivate endogenous peroxidases (Perhydrol, 30% per analysis; Merck, Darmstadt, Germany). Differences in assay for the individual antibodies are indicated below by superscripts, with 1 for RBP, 2 for CRABP I, and 3 for RXRß. The superscript A stands for dilution in 20% swine serum in Tris-buffered saline (TBS), and superscript B for a dilution in 1% crystalline BSA (Boehringer Mannheim, Mannheim, Germany) in TBS. Nonspecific antibody binding was blocked for 10 min in TBS, pH 7.6, containing 50% normal porcine1/rat serum2,3 and incubated with the primary antibody (RBP, 1:400A; CRABP I, 1:300B; RXRß, 1:150B). Primary antibody incubations were performed in a humidified chamber at 4°C overnight. The sections were then incubated with swine anti-rabbit IgG1 (1:100A)/rat anti-mouse IgG antibody2,3 (1:100B) for 30 min. Finally the sections were incubated for 30 min with rabbit PAP1 (1:100A)/mouse PAP2,3 (1:100B). Sections were washed several times with TBS, pH 7.6, between each incubation. Diaminobenzidine and 0.01% H2O2 (30%) were used in 60 mM imidazole buffer, pH 7.1, as substrates for the peroxidase to produce the brown stain that revealed the presence of immunoreactive substances in the sections. The sections were counterstained with Papanicolaou (Merck), then dehydrated and covered with a coverslip. In control sections, normal serum was used instead of the primary antibody. All incubations without the primary antibody were performed at room temperature.
Analysis of Vitamin A
Vitamin A was extracted from uterine tissue (0.61.0 g) using n-hexane and isopropanol (3:2; v:v, 0.05 butylated hydroxytoluene) to homogenize the samples [11]. The combined organic phases were dried under nitrogen and redissolved in methanol:ethanol (4:1 v:v). Vitamin A (retinol, retinyl esters) was separated and quantified as described previously with slight modifications [12] using a reversed-phase HPLC system (Waters, Eschborn, Germany) on an RP-18 column (5 µm, 125 x 4 mm; Grom, Ammerbuch, Germany) with methanol as solvent at a flow rate of 1 ml (13 min), 1.5 ml (37 min), and 2 ml (719 min). Identification and quantification of vitamins were obtained by comparison of retention time as well as peak areas at 325 nm with external standards (Serva, Heidelberg, Germany). Data for retinyl esters are presented as retinol equivalents [13]. All solvents or chemicals used were of high-purity commercial grade (Merck).
Statistical Analysis
Data are reported as means ± SEM. Data of myometrium and endometrium were compared by Student's t-test. Variations in response variables (retinol and retinyl esters) were partitioned using ANOVA procedure of SAS (Version 7.0; SAS Institute, Cary, NC). If a significant difference was found, a Scheffé post hoc test was used to determine the cause of the significant difference. The probability level at which differences were considered significant was P < 0.05.
| RESULTS |
|---|
|
|
|---|
|
|
The fluorescence microscopic investigation of cryosections showed a typical yellowish-greenish fluorescence at all investigated implantation sites and intersegments, with no obvious differences in fluorescence intensity on any given day of pregnancy. However, there were differences depending on the progress of gestation. During the early days, the fluorescence was more intensive than later on. On Day 30, more and more areas without any fluorescence were observed. As shown in Figure 1, c and d, the typical fluorescence was limited to the endometrium of the uterus and was associated with the uterine glands and the uterine surface epithelium. The fluorescence faded very quickly (< 30 sec) under exposure to UV light. Furthermore, no fluorescence was observed when cryosections were treated with the organic solvent n-hexane, which removes the lipophilic components such as retinol and retinyl palmitate, prior to fluorescence microscopy.
When total uterine tissue on Day 28 of gestation was selected according to fluorescence and nonfluorescence areas, the amount of total vitamin A in regions with fluorescence (484.3 ± 56.2 ng/g tissue wet weight) was significantly higher (P < 0.001) than in samples with no fluorescence (125.8 ± 9.3 ng/g; mean ± SEM, 5 samples each of different locations from 4 animals).
Levels of retinol and retinyl esters in the endometrium and myometrium of the uterus at different days of gestation are summarized in Table 1 and Table 2, respectively. Since no significant differences were observed between the various uterine sites from which samples were obtained, results were combined for the individual days of gestation. Results show that at all time points, the level of total vitamin A (retinol and retinyl esters) in the endometrium was higher than in the myometrium (P < 0.01). In both the endometrium and myometrium, total vitamin A showed a tendency to decrease with time. On Day 30 of gestation, no vitamin A could be detected in the myometrium. The most obvious difference between myometrium and endometrium in the uterus was that in the endometrium, the dominant fraction of vitamin A was almost exclusively retinol, 89.8 ± 1.7% (mean ± SEM of all samples; n = 60). Retinyl esters were predominantly present as retinyl palmitate and traces of retinyl stearate. In the myometrium, however, retinol represented only 8.5 ± 3.9%, and retinyl esters, predominantly as retinyl palmitate, represented the majority of total vitamin A (mean ± SEM of all samples, n = 62). Differences in retinol (P < 0.001) and total retinyl esters (P < 0.01) between all endometrium and myometrium samples were significant (Student's t-test).
|
|
Immunoreactive RBP was detected throughout early gestation in all sections examined and was always limited to the endometrium (Fig. 1e). Early in gestation, RBP was present in the glandular and surface epithelium, but on Day 30, RBP was predominantly present in the surface epithelium.
Similar to RBP, CRABP I in the uterus was always limited to the endometrial cells. At Day 14, CRABP I was found more evenly distributed throughout the cytoplasm of endometrial cells. At Days 16, 18, and 20, CRABP I was more oriented toward the apical region of the cells (Fig. 1f) and more prominent in the surface compared to the glandular epithelium. On later days, immunoreactive CRABP I was again more evenly distributed in the cytoplasm and was found in comparable intensities in surface and glandular epithelium. At Day 30, general intensity decreased again and was limited to the glandular epithelium.
RXRß was found at all days of gestation investigated. The nuclear receptor was strictly limited to the cell nucleus. Greatest intensity was observed at Days 14 and 16 in both the surface and the glandular epithelium. On later days, the intensity was reduced and was limited to the glandular epithelium (Fig. 1g). The presence of RXRß in the myometrium was independent of the day of gestation (Fig. 1h).
| DISCUSSION |
|---|
|
|
|---|
Both HPLC analysis and the microscopic visualization of vitamin A under UV light showed that vitamin A is not equally distributed in uterine tissue but is enriched in the endometrium. Fluorescence microscopy further showed that the autofluorescence was not uniformly distributed in uterine tissue but was limited to the glandular and surface epithelium of the endometrium. The distribution of fluorescence was similar to the distribution of immunoreactive RBP. This indicates an accumulation of vitamin A in cells where the synthesis and secretion of RBP took place. Since nearly all of the vitamin A in the endometrium was present as retinol, this strongly points to the importance of endometrial vitamin A as a source for the secretion of RBP (holo-RBP)-associated vitamin A into the uterine lumen to supply the developing conceptus with vitamin A prior to placentation.
Different lines of evidence support the hypothesis that the yellowish-greenish fluorescence described in uterine tissue was caused by vitamin A [5]. First, the UV-excitation wavelength and color of the fluorescence are typical for vitamin A [7]. Second, the autofluorescence in the uterus disappeared within a very short time of UV exposurea fact that has been described for vitamin A but for no other substances showing autofluorescence in human or animal tissue [6, 16]. Third, the fluorescence was not observed when the cryosections were treated with n-hexane prior to UV exposure, indicating that the lipophilic vitamin A is extracted into the organic solvent. This was verified by the detection of vitamin A in the solvent after incubation of the sections (data not shown). Fourth, substantial differences in vitamin A content of uterine tissue were observed when tissue samples showing autofluorescence were compared with those having no fluorescence. Finally, the specific area showed an emission maximum of 476 nm, which corresponds to the in vitro emission maxima of retinol of 475510 nm depending on the solvent [16].
Numerous studies using in situ hybridization and immunocytochemistry have recently shown that uterine glands and uterine surface epithelium of the pig and of other species are the sites of RBP synthesis [1721]. Our results and those of others show that in pigs on Day 14 of gestation, the uterine gland and uterine surface epithelium contain RBP, while on Day 30, RBP is limited to the surface epithelium [22]. In the myometrium, immunoreactive RBP was present only on Day 60 [22]. The RBP synthesized in uterine endometrium is secreted together with retinol into the uterine lumen in response to progesterone; it constitutes one of the major secretory proteins in the uterine histotroph [2, 8, 17]. Its function is presumably to transport and deliver retinol from the uterus to the developing conceptus before implantation [23, 24]. Additionally, the porcine conceptus itself starts to produce and secrete RBP as early as Day 10 [25]. In the uterine fluid, the histotroph, RBP is present as apo- and holo-RBP. The origin of the retinol bound to RBP has not been investigated to date, but as our results show, it is very probably derived from the uterus, where it is taken up from plasma and temporarily accumulated during early gestation.
Besides being secreted into the uterine lumen, vitamin A might serve as a local source of retinoic acid. Increased levels of retinoic acid and CRABP II have been measured in the uteri of prepubescent rats treated with eCG [26]. Cellular binding proteins for retinol (CRBP) and retinoic acid (CRABP) were observed in the smooth muscle layers of the uterus. In these structures, CRABP might serve to protect muscle cells from the effects of retinoic acid produced around estrus by the uterine epithelium, or may contribute to the general metabolism of retinoic acid in these cells [27]. The continuous decrease in total vitamin A with the progress of gestation in uterine tissue might point to the fact that later on in gestation the vitamin A supply to the conceptus is regulated via the transfer through the placenta. So far, it is not known whether vitamin A is transferred from the myometrium to the endometrium or whether it is directly derived from plasma. The first alternative would be supported by the observation that vitamin A in the myometrium is mainly present in its storage form, the retinyl esters, thus possibly involving the local esterification of retinol and/or the hydrolysis of retinyl esters. The participation of the myometrium in the storage of vitamin A has been discussed in connection with the occurrence of CRBP in rat myometrium [27] because retinol uptake and subsequent esterification are mediated by CRBP [28]. On the other hand, the localization of vitamin A shown by macroscopic autofluorescence might favor its accumulation directly from the blood because in these areas an increased uterine vascular permeability can be observed, which might facilitate the transcapillary traffic of precursors for histotroph production [29]. But myometrial vitamin A might not only be a depot for endometrial secretion activity; it might also be of local importance for the myometrium itself, as we were able to show that porcine myometrium contains RXRß at all time points investigated. This finding corresponds to results in humans, in which uterine smooth muscle cells and endometrium express retinoic acid receptors RAR
, RARß, and RAR
, as well as RXR
and RXRß, throughout the menstrual cycle [30]. Estrogen-induced uterine stroma and myometrial cell proliferation is inhibited in a dose-dependent manner by all-trans retinoic acid [31, 32]. Because of the analytical system used, retinoic acid was not investigated in our study.
In conclusion, this study presents for the first time results on the content and the localization of vitamin A in the uterus of pigs, showing that the yellowish-greenish autofluorescence in the pig uterus can be attributed to vitamin A. Despite some limitations arising mainly from the instability of vitamin A when exposed to UV light, causing problems in localizing the vitamin A precisely in tissue if present at lower concentrations and in more complex structures, the macroscopic and microscopic investigation of the autofluorescence of vitamin A provided additional power to study the distribution of vitamin A at a morphological level. The combination of fluorescence microscopy and immunological techniques showed that both vitamin A and immunoreactive RBP were present in the same structures: the secretory epithelium of the uterus. Differences in the form in which vitamin A is present in endometrium and myometrium might point to differences in metabolism of vitamin A with regard to storage and secretion in the porcine uterus.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
2 Correspondence: Florian J. Schweigert, Institute of Nutritional Science, University Potsdam, Arthur-Scheunert-Allee 114-116, D-14558 Bergholz-Rehbrücke, Germany. FAX: 033200 88573; fjschwei{at}rz.uni-potsdam.de ![]()
Accepted: May 6, 1999.
Received: August 14, 1998.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Dupuis, A. Levy, and S. Mhaouty-Kodja Functional Coupling of Rat Myometrial {alpha}1-Adrenergic Receptors to Gh{alpha}/Tissue Transglutaminase 2 during Pregnancy J. Biol. Chem., April 30, 2004; 279(18): 19257 - 19263. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Johansson, L. Dencker, and V. Dantzer Immunohistochemical Localization of Retinoid Binding Proteins at the Materno-Fetal Interface of the Porcine Epitheliochorial Placenta Biol Reprod, January 1, 2001; 64(1): 60 - 68. [Abstract] [Full Text] |
||||
![]() |
J. Vermot, V. Fraulob, P. Dolle, and K. Niederreither Expression of Enzymes Synthesizing (Aldehyde Dehydrogenase 1 and Retinaldehyde Dehydrogenase 2) and Metabolizing (Cyp26) Retinoic Acid in the Mouse Female Reproductive System Endocrinology, October 1, 2000; 141(10): 3638 - 3645. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |