Human Reproduction vol.7 no.9 pp.1214-1221, 1992
Cytokine expression in human endometrium throughout the menstrual cycle
S.Tabibzadeh1 and X.Z.Sun Department of Pathology, University of South Florida Health Sciences Center and Moffitt Cancer Center, Tampa, FL 33612, USA 'To whom correspondence should be addressed at: Department of Pathology, Moffitt Cancer Center, 12902 Magnolia Drive, Tampa, FL 33612, USA
Recent evidence suggests that diverse endometrial functions may be regulated by cytokines. In this report, the presence of protein and mRNA of cytokines were studied in human endometrium throughout the menstrual cycle. The presence of the interkukin-1 (1L-1)«, interieukin-1 (TL-l)jS, interieukin receptor antagonist (TRAP), interleukin-6 (IL-6) and transforming growth factor (TGF>a proteins were demonstrated by immunohistochemkal staining. The IL-la and TGF-a proteins were strongly expressed and IL-1/3 protein was weakly expressed in all the cells in the stroma as well as epithelial cells. IRAP was markedly expressed in the cells with morphological features of macrophages scattered in the stroma, and the expression of IL-6 protein was predominant in the endometrial epithelium. Diffuse cytoplasmic expression of IL-la in endometrial epithelium during the proliferative phase contrasted markedly with its enhanced luminal expression during the secretory phase of the menstrual cycle. In addition, the presence of the mRNA of these cytokines in endometrium was established throughout the entire menstrual cyde by reverse transcription-polymerase chain reaction (RT-PCR). Abundant expression of cytokines in human endometrium emphasizes the significant roles that cytokines play in cell-cell interactions and in regulating endometrial functions. Key words: cytokine/endometrium/growth polymerase chain reaction
factor/mRNA/
whether some of the effects of tiiese hormones are effects directed at the target tissue or are indirectly exerted through elaboration of other factors within the endometrium. It is conceivable that many cell-cell interactions and steroid-cell interactions in the endometrium occur via the production of factors that are collectively called cytokines. The emerging evidence suggests that many of the endometrial functions are controlled by these factors. Such diverse functions as proliferation, secretion of prostaglandin E2 (PGE2) or cytokines, human leukocyte antigen (HLA)-DR expression, and differentiation may be regulated by various cytokines (Tabibzadeh, 1991a; Tabibzadeh et al., 1986, 1988, 1990a,b; Karyia et al., 1991). In addition, these factors may be under the controlling influence of steroid hormones linking the effect of local factors to die systemic signals (Pacifici et al., 1989; Tabibzadeh et al., 1989; Ralston et al., 1990; Fox et al., 1991). In view of the overwhelming evidence for the contribution of the cytokines in the regulation of endometrial functions, this study was designed to demonstrate the expression of the proteins and the genes of several cytokines in human endometrium. Reverse transcription followed by polymerase chain reaction (RT-PCR) has emerged as an extremely sensitive and rapid technique which allows detection of low copy-number mRNA species including those of cytokines. In studies where the amount of available tissue for isolation of RNA is limited, this technique allows detection of the expression of multiple genes, a technique called mRNA phenotyping (Rapopolee et al., 1988). In view of our interest to show the presence of mRNA of several cytokines in small amounts of endometria, and particularly to verify their expression throughout the entire menstrual cycle, this technique seemed suitable. Subsequently, the expression of cytokine proteins was studied in situ by an immunohistochemical staining procedure. Materials and methods
Introduction Human endometrium is composed of glandular structures which are invested with endometrial stroma. Both components characteristically undergo a predictable series of structural, morphological, cytochemical and immunohistochemical changes (Tabibzadeh, 1990, 1991a). These changes are accompanied by sequences of proliferation (proliferative phase), secretion (secretory phase) and menstrual shedding (menstruation). It is thought that these changes are primarily driven by two steroid hormones, oestrogen and progesterone. However, it is unclear 1214
The mouse monoclonal antibodies to interieukin-la (IL-la), interieukin-1/3 (IL-1/3), and transforming growth factor-a (TGFa) were obtained from Olympus Immunochemicals (Lake Success, NY, USA). The monoclonal antibody to interieukin receptor antagonist protein (IRAP) was a generous gift of Dr M.J.Bienkowski (Upjohn Co., Kalamazoo, Michigan, USA). The IgG fraction of a polyclonal rabbit antiserum to interleukin-6 (IL-6) was kindly provided by Dr L.May (Rockefeller University, NY, USA). Primers to the IRAP were kindly provided by Dr D.B.Carter (Upjohn Co.). All other primers used in tius study were obtained from Clontech (Palo Alto, CA, USA). The © Oxford University Press
Cytokines in endometrium
TaWe I. Primers used in the polymerase chain reactions (PCR) and the size of the amplified products Name
Sequences of 5' and 3' primers
IL-la
5' ATGGCCAAAGTTCCAGACATGTTTG 3' GGTTTTCCAGTATCTGAAAGTCAGT
IL-10
IRAP
IL-6
TGF-a
0-Actin
Size of amplified products (bp)
5' ATGGCAGAAGTACCTAAGCTCGC 3' ACACAAATTGCATGGTGAAGTCAGTT
5' CCAGAAGACCTTCTATCTGAGGA 3' CCAGCTGGAGGCAGTTAAGATCAC
816 (Hireffll: 190;626) 802 (Hindm: 396;406) {TaqY. 314;488) (Nco\: 209;593) 201 (NcoY. 77; 124) (Sau96l: 53,148)
5' ATGAACTCCTTCTCCACAAGCGC 3' GAAGAGCCCTCAGGCTGGACTG (Taql: (XbaY.
628 182;446) 148;480)
(Rsal:
297 69;228)
5' ATGGTCCCCTCGGCTGGACAG 3' GGCCTGCTTCTTCTGGCTGGCA 5' ATGGATGATGATATCGCCGCGG 3' CTAGAAGCATTTGCGGTGGACGATGGAGGGGCC
1126
The lengths of the restriction enzyme-digested products (in bp) are indicated in parentheses. IL = interleukin; IRAP = interleuldn receptor antagonist protein; TGF = transforming growth factor.
sequences of the 5' and 3' primers and the lengths (in base pairs, bp) of the amplified products from RT-PCR reactions of cytokine mRNA are shown in Table I. The AMV reverse transcriptase was obtained from Boehringer Mannheim (Emeryville, CA, USA) and the Amplitaq was purchased from Cetus (Norwalk, CT, USA). Processing of endometria Endometria (n = 31) were from hysterectomy specimens from patients in their reproductive years. The uteri were removed for abnormalities which were not endometrial in origin, including uterine leiomyomas, and cervical or ovarian lesions. Upon removal, uteri were quickly transferred to the laboratory. Blocks of endometrium, 0.5 cm in diameter, were removed and embedded in paraffin. Sections were stained by haematoxylin and eosin and dated according to the published criteria (Noyes and Hertig, 1955). Identical blocks were placed in OCT compound, then snap frozen in a mixture of methylpropane and dry ice and kept at -70°C prior to immunostaining. Remaining endometrium was frozen and kept in liquid nitrogen for isolation of RNA. Endometria were dated with respect to early proliferative (n = 2), mid-proliferative (n = 3), late proliferative (n = 5), early secretory (n = 6), mid-secretory (n = 8) and late secretory (n = 7) phases of the menstrual cycle. Endometria used in this study showed normal morphology. Immunohistochemical staining procedure Air dried, 4-/tm-thick cryostat sections were immunostained by the avidin—biotin complex (ABC) staining method (Hsu et al., 1981; Tabibzadeh and Gerber, 1986). Briefly, the immunostaining procedure consisted of incubation with normal horse serum, mouse monoclonal or rabbit polyclonal antibodies, the appropriate biotinylated horse anti-mouse or biotinylated goat
anti-rabbit antibodies, ABC complexes and then development in a mixture of 4-chloro-l-naphthol (CN)-H2O2, resulting in a blue reaction product. The polyclonal antibody was used at 1/100 dilution and the monoclonal antibodies were used at 1 - 2 /*g/ml. Secondary antibodies were used at 1/200 dilution. Negative controls for immunostaining consisted of the substitution of primary antibody by irrelevant antibody, normal rabbit serum, rabbit or mouse immunoglobulins at equivalent protein concentrations or phosphate-buffered saline (PBS). Endogenous avidinbinding activity and endogenous peroxidase activity were demonstrated respectively by incubation with avidin-biotin complex and then development in a mixture of 4-chloro-lnaphthol (CNH^Qz or directly by incubation with the substrate. Alternatively, the endogenous peroxidase was quenched during the immunostaining procedure for 30 min in 0.3% H2C>2 in methanol following incubation of the section with the primary and then the biotinylated antibodies and prior to application of the ABC complexes. The tissues were examined without counterstaining. The intensity of the positive reaction with the immunostaining was semi-quantitatively expressed as: strong, moderate, weak and no staining. Isolation of RNA Total RNA was isolated by the guanidinium isothiocyanatecaesium chloride method from endometrial curettings weighing 20-100 mg (Farrell, 1989; Sambrook et al., 1989). Following digestion in DNase, a 15 /tg aliquot of the RNA was electrophoresed on formaldehyde denaturing gel to detect the intact 18S and 28S ribosomal RNA and to confirm the integrity of the endometrial RNA. The amount of RNA was calculated by measuring optical density at 260 nm in a spectrophotometer (Perkin-Elmer). 1215
S.Tabibzadeh and X.Z.Sun
Restriction enzyme digestion of amplified products
Table D. Cytoldne profiles in the endometria used in the reverse transcriptase—polymerase chain reaction (RT—PCR) IL-1 alpha
IRAP
IL-6
TGF-a
ND ND
ND ND
ND ND
ND ND
ND
ND
+ +-
Mid-P Mid-P Late-P Early-S Early-S Mid-S Mid-S Mid-S Mid-S Late-S Late-S
IL-1 beta
+
ND
ND
ND
+ +
Date of endometrium
ND
Abbreviations: P, proliferative; S, secretory; + , positive; ND, not done (insufficient amount of tissue); IL = interleukin; IRAP = interleultin receptor antagonist protein; TGF = transforming growth factor.
Reverse transcription Reverse transcription was performed as described (Rapopolee et al., 1988; Sambrook et al., 1989). Following heating at 65 °C for 5 min, 1 /ig aliquot of total RNA in 7.5 /J of diethylpyrocarbonate (DEPC)-treated water was used in each reverse transcription reaction. Reverse transcription was allowed in the presence of 5 units of AMV reverse transcriptase, 40 units of RNAsin, 5 /xl of 5 X reverse transcription buffer (500 mM Tris-HCl, pH 8.3, 250 mM KC1, 50 mM MgCl2, 50 mM dithiothreitol), 2.5 /J of random hexamer (0.5 /tg//J), 2.5 /d dNTP mixture (10 mM each). The total reaction volume was 17.5 id and the reaction was allowed to continue at 42°C for 1 h. Polymerase chain reaction The reverse transcription reaction mixture was used in a 100 /il PCR as follows. PCR was carried out by the addition of Amplitaq (2.5 IU), dNTP (200 /iM), primers (1 /*M), 10 /il of lOx reaction buffer (100 mM Tris-HCl, 500 mM KC1, 15 mM MgCl2, 0.01% gelatin) and water. The PCR reaction mixture was overlaid with mineral oil and amplification was carried out in a thermocycler (Perkin-Elmer/Cetus, Norwalk, CT, USA). The 30-cycle amplification profile consisted of denaturation at 95 °C for 1 min, annealing at 55°C for 2 min, extension at 72°C for 2 min with a final extension at 72°C for 5 min. The thermocycler was located distant from the preparation area and the amplified products were analysed in a separate laboratory. Fractions (10-20 /xl) of the amplified products along with *xl74 RF DNAJHaeUl fragments (1353, 1078, 872, 603, 310, 271; 281, 234, 194, 118, and 72 bp) serving as molecular weight standards were electrophoresed on a 2—2.5% Nusieve-agarose gel containing ethidium bromide and the bands were visualized and photographed in ultraviolet light. In each PCR, the RNA was amplified without the reverse transcription step to detect the presence of any contaminating genomic DNA. Also, a tube containing all the PCR components except the RT reaction mixture served as a negative control to check for the presence of DNA that may have been carried over from a prior reaction. Endometria which were used in the RT reactions are listed in Table n . 1216
The expected lengths (in bp) of the restriction enzyme-digested products of amplicons of cytokines were determined using a recombinant toolkit (Biosoft, Cambridge, UK). The band containing the amplified product from 200—300 /d PCR reactions was removed following electrophoresis and then extracted with phenol-chloroform and precipitated in a mixture of 100% ethanol and 1/10 volume of 3 mM sodium acetate. Following the enzyme digestions, at appropriate temperature for 30 min to 2 h, the samples were electrophoresed to determine the lengths of the digested products. The expected lengths (in bp) of the amplified products of the cytokines as well as their respective restriction enzyme-digested products are shown in Table I.
Results Cellular localization of cytoldne proteins The cytokine proteins could be localized in all the cases studied. Epithelial cells and all the cells in the stroma, including stromal, lymphoid and endothelial cells, expressed all these cytokines; however, the intensity of staining in cells of different lineages was different with respect to each specific cytokine. IL-1 a and TGF-a were strongly stained in equal intensity in cells of different lineages both in the epithelium and stroma and the endothelial cells (Figures 1-3, Table HI). The distribution pattern of immunostaining for IL-1/3 was similar to the staining profile of IL-la; however, the intensity of the immunostaining was weak to absent (Figure 4, Table IU). The strongest staining of IRAP was observed in cells with the morphological characteristics of macrophages, such as abundant cytoplasm and bean-shaped nuclei. These cells were scattered in the endometrial stroma (Figure 5, Table HI). The epithelial, stromal and endothelial cells exhibited weak to moderate staining for IRAP (Figure 5). The epithelial cells exhibited marked staining and the cells in the stroma showed weak to moderate staining for IL-6 (Figure 6, Table III). Some regional and gland—gland variations could be observed in the cases which were immunostained for cytokines. In control sections, no reaction product was present upon omission of the primary antibodies and their substitution with irrelevant antibodies, PBS or animal serum. However, endogenous peroxidase activity was observed in polymorphonuclear leukocytes and macrophages and pseudoperoxidase activity was present in the red blood cells in specifically stained sections as well as controls (Figure 6). Under the experimental conditions, no endogenous avidin-binding activity was observed in any of the cases studied.
Temporal changes in the expression of cytokine proteins Some menstrual cycle-related changes could be observed in the expression of the cytokine proteins. This was most evident in the expression of IL-la. The immunostaining for IL-la was rather uniformly expressed in the cytoplasm of the endometrial glands in the proliferative phase (Figure 1). In the mid- to late secretory phases, however, the immunoreactivity for IL-la shifted luminally (Figure 2).
Cytoldnes in endometriuro
FTg. 1. Immunoreactivity for interieukin-1 (IL-l)a is strongly expressed in the epithelium and the stroma in a late proliferan've endometrium (X200). Fig. 3. Immunoreactivity of the transforming growth factor (TGF-a) is strongly observed in cells in the stroma (small arrowheads), endothelial cells (large arrowheads) and epithelium (arrows) in this mid-proliferative endometrium (X160).
TaMe HI. Differential expression of cytokine proteins in epithelium versus stroma in human endometrium Intensity of staining
IL-la IL-1/3 IRAP TGF-a IL-6
Fig. 2. Immunoreactivity for interleukin-1 (IL-l)ar is observed in the luminal bonders of the epithelium in a mid-secretory endometrium (large arrowheads). Small arrowheads point to the positively immunostained cells in the stroma (xl60).
Epithelium
Stroma
Strong Weak Moderate Strong Strong
Strong Weak Weak (strong expression in macrophages) Strong Weak to moderate
IL = interieukin; IRAP = interieukin receptor antagonist protein; TGF = transforming growth factor.
i, .
Demonstration of cytotdne gene expression The IL-la, IL-l/S, IRAP, IL-6 and TGF-a gene expression was demonstrated by the reverse transcription of their respective mRNA to cDNA using random hexamers as primers for the AMV reverse transcriptase-mediated reaction followed by amplification using cytokine-specific primers (Table I). The presence of amplified products was detected by their mobility after electrophoresis in ethidium bromide-stained agarose gels and further identification of the amplified products was made by restriction enzyme digestion. Amplicons of all cytokines were observed in all the cases studied. The amplicons of amplified products from RT-PCR reactions for various cytokines were of the expected lengths (bp) (Table I, Figures 7-12). The lengths (in bp) of the restriction enzyme-digested products also conformed to those expected (Table I, Figures 7 — 12). Non-specific products were not observed in any of the RT—PCR reactions (Figures 7—12). The amplification of the /3-actin mRNA served as a positive control in all the RT-PCR reactions. In all reactions, and in all the cases studied, the amplified product for /3-actin
Fig. 4. Immunoreacu'vity for interleukin-1 (IL-l)/3 is weakly expressed in epithelium and the stroma in this proliferative endometrium (X100).
could be observed. Contaminating genomic DNA was not detected in PCR, as the total RNA amplified without the reverse transcription step did notrevealany amplified product (Figure 7). 1217
S.TaMbzaddi and X.Z.Sun
Fig. 5. The strongest expression of immunoreactive interleukin receptor antagonist protein (IRAP) is observed in some cells scattered in the stroma with morphological features of macrophages (large arrowheads). Epithelial cells (arrow) exhibit a moderate staining and cells in the stroma (small arrowhead) a weak staining in this early proliferative endometrium (X160).
Fig. 7. Electrophoresis of the products of the reverse transcription—polymerase chain reaction (RT—PCR) of interleukin (EL)-la, EL-1/3 and interleukin receptor antagonist protein (IRAP) mRNA of endometrium. Lane 1: amplification of mRNA without reverse transcription step. Lane 2: amplification product of EL-la mRNA (816 bp). Lane 3: amplification of mRNA without reverse transcription. Lane 4: amplified product of EL-1/3 mRNA (802 bp). Lane 5: xl74 replication factor (RF) DNAJHaeUl fragments. Lane 6: amplified product of IRAP mRNA (201 bp).
Fig. 6. Immunoreactive interleukin-(IL)-6 is strongly expressed in the glandular epithelial cells in this mid-secretory endometrium. The arrowhead points to an endogenous peroxidase positive cell (X160).
In addition, amplified product was not observed in PCR reactions where all the PCR components were used except for the addition of the RT reaction mixture.
Discussion In this report, the presence and distribution of cytokine-positive cells was shown by immunostaining of sections of endometria. The presence of the cytokine mRNAs was confirmed by RT—PCR and cells of different lineages expressed cytokine proteins. They included epithelial, endothelial, lymphoid and stromal cells. IL-la and TGF-a showed the widest distribution with respect to the cell types, the regional and temporal distributions. IL-1/3 showed a weak but similar distribution. This is in 1218
Fig. 8. Electrophoresis of the products of the reverse transcription—polymerase chain reaction (RT-PCR) of interleukin (IL)-la and IL-1/3 mRNA of endometrium. Lane 1: amplified product of IL-la mRNA (816 bp). Lane 2: amplified product of IL-la digested with MndHI (190 and 626 bp). Lane 3: amplified product of EL-1/3 mRNA (802 bp). Lane 4: The 396 and 406 bp products of Hindm digestion of the amplified product of IL-1/3 are overlapped. Lane 5: *xl74 RF DNA/WaeHI fragments.
Cytoklnes In endotnetrtuin
Fig. 9. Electrophoresis of the products of the reverse transcription—polymerase chain reaction (RT—PCR) of interieukin (IL)-/3 mRNA of endometrium. Lane 1: blank. Lane 2: amplified product of IL-1/9 mRNA (802 bp). Lane 3: amplified product of IL-l/S digested with Taql (314;488 bp). Lane 4: amplified product of IL-10 mRNA digested with Ncol (209;593 bp). Lane 5: xl74 RF DfiAJHaem fragments.
FTg. 11. Electrophoresis of the product of the reverse transcription-polymerase chain reaction (RT—PCR) of transforming growth factor (TGF)-a mRNA of endometrium. Lane 1: amplified product of TGF-a mRNA (297 bp). Lane 2: amplified product of TGF-a digested with Rsal (69 and 228 bp). Lane 3: *xl74 RF DNAJHaeUl fragments.
Fig. 10. Electrophoresis of the product of the reverse transcription—polymerase chain reaction (RT—PCR) of interieukin receptor antagonist protein (IRAP) mRNA of endometrium. Lane 1: amplified product of IRAP mRNA (201 bp). Lane 2: amplified product of IRAP digested with Ncol (77 and 124 bp). Lane 3: amplified product of IRAP mRNA digested with Sau96I (148;53 bp). Lane 4: *xl74 RF DNA//faem fragments.
F ^ . 12. Electrophoresis of the product of the reverse transcription—polymerase chain reaction (RT—PCR) of interieukin (IL)-6 mRNA of endometrium. Lane 1: *xl74 RF DtiAJHaeTa fragments. Lanes 2 and 5: amplified products of IL-6 mRNA (628 bp). Lane 3: amplified product of IL-6 digested with Xba\ (148 and 480 bp). Some undigested product has remained. Lane 4: blank. Lane 6: amplified product of IL-6 digested with Taql (182 and 446 bp).
line with the evidence that the mRNA of IL-1/3 is not observed by Northern blotting in the proliferative phase and is only rarely detected in the secretory phase in human endometria (Kauma et at., 1990). In contrast, IL-1/3 mRNA is easily detected in the gestational endometria and IL-1 activity may be detected in the human decidua and in the amniotic fluid (Romero et at., 1989a; Tamatani et at., 1988). In contrast to our finding of the wide distribution of IL-1 a and IL-1/3 in human endometrium, the message for IL-1 is apparently limited to cells with the
morphological characteristics of macrophages in mouse uteri (Takacs et at., 1985). Temporal and cell type-specific expression of the TGF-a transcript and protein have also been demonstrated in mouse uteri in the peri-implantation period by immunohistochemistry as well as by in-situ and Northern blot hybridizations (Lee, 1990; Tamada et at., 1991). This expression was shown both in the luminal and glandular epithelia on days 1 —4 and in many stromal cells on days 3 and 4 of pregnancy. The present 1219
S.TaWbzadeh and X.Z.Sun
study shows that in contrast to the equal intensity of immunoreactivity of IL-la and TGF-a in both epithelial and stromal cell components, IRAP is strongly expressed in cells in the stroma with morphological characteristics of macrophages whereas IL-6 is strongly expressed in the epithelial cells. Menstrual cycle-related changes in the expression of most cytokine proteins was not apparent This could be due to the semiquantitative nature of the immunostaining procedure. However, a menstrual cycle-related change in the expression of IL-la gene product was observed in the endometrial epithelium; the diffuse cytoplasmic staining in the proliferative phase shifted luminally in the secretory phase. This menstrual cycle-related change in the expression of a cytokine in human endometrium is not unprecedented. We have observed immunoreactivity for TNF-a in endometrial epithelial cells primarily in the secretory phase of the menstrual cycle. In addition, the immunoreactivity of TNF-a is strongest in stromal cells in the proliferative phase and is diminished in the secretory phase (Tabibzadeh, 1991b; Hunt et al, 1992). Also, the expression of the colony-stimulating factor (CSF)-1 transcripts in endometrial epithelium shows a distinct menstrual cycle-related pattern (Pampfer et al., 1991). Endometrium is extremely sensitive to steroid action and the sequential morphological, functional and immunochemical profiles of this tissue are primarily driven by the actions of steroids (Tabibzadeh, 1991a). Thus, it is conceivable that the menstrual cycle-related changes in the expression of cytokines in human endometrium also have a similar basis, hi accordance with this hypothesis is the demonstration of the blockage of release of both IL-1 and TNF-a by gonadal steroid treatment in monocytes (Pacifici et al., 1989; Ralston et al., 1990). Similarly, we have shown that the IL-1 mediated expression of IL-6 in human endometrial stromal cells may be down-regulated by oestrogen (Tabibzadeh et al., 1989). TGF-a transcript which is not detectable in non-pregnant mouse uteri, can be detected in oestrogen-treated uteri and high levels (50-70 ng/ml) of TGF-a protein are observed by Western blotting after diethylstilbestrol (DES) stimulation (Lee, 1990; Tamada et al, 1991). A recent study indicates that the 5'-flanking region of the TGF-a contains a potential oestrogen-responsive element(s) (Saeki et al., 1992). The ability of the oestradiol to regulate the interferon (IFN)gamma promoter and to increase the IFN-gamma mRNA expression in Con A-activated murine spleen cells provides further proof of interaction of steroid hormone with cytokines (Fox et al., 1991). Taken together, these data are in accordance with the hypothesis that steroid hormones may alter the expression of cytokines in endometrium and that some of the effects of oestrogen may be mediated through cytokines in this tissue. For example, it has been suggested that the mitogenic effect of oestradiol in endometrium may be mediated through epidermal growth factor (EGF)/TGF-a (Lee, 1990). With the recognition that many of the cell—cell interactions in endometrium may be driven by cytokines, diverse functions for the cytokines in endometrium are being described. These include control of growth, differentiation, secretion of prostaglandin-E2 and HLA-DR expression (Karyia et al., 1991; Tabibzadeh, 1991a; Tabibzadeh et al., 1986, 1988, 1989, 1990a,b). We have observed that the synthesis of IL-6 by the endometrial stromal cells may be induced by IL-1 as well as by 1220
IFN-gamma and TNF-a (Tabibzadeh et al., 1989). IL-1 also induces the synthesis of PGE2 in endometrial epithelial cells (Tabibzadeh et al., 1990b). Whereas progesterone drives the human endometrial stromal cells to become decidualized in vitro, IL-1 inhibits such differentiation (Karyia et al., 1991). IL-1 with its potent biological activity, is likely to be under stringent regulatory control in vivo. IRAP has now been recognized as one such regulatory mechanism (Carter et al., 1990; Hannum et al, 1990; Arend, 1991). IRAP blocks multiple biological activities of IL-1 including release of PGE2 by fibroblasts, synovia! cells and-chondrocytes, thymocyte proliferation, and collagenase production by synovial cells (Carter et al., 1990; Hannum etal., 1990; Arend et al., 1990; Arend, 1991; Mclntyre et al., 1991). Thus, IRAP may function in a similar manner in human endometrium to control the multiple actions of IL-1. The enhanced production of the placental IL-1 has been shown during labour and in intra-uterine infection and it has been suggested that this cytokine is involved in parturition (Romero et al., 1989a). It has also been speculated that TGF-a, IL-6, TNF-a, M-CSF (macrophage-colony stimulating factor) and GM-CSF (granulocyte-macrophage colony stimulating factor) are involved in the reproductive functions (Romero et al., 1989b, 1990; Tamada etal., 1991; Wegmann et al., 1989). In summary, we have demonstrated the expression of several cytokine genes and their respective products in human endometrium throughout the menstrual cycle. Cytokines were characterized by their cell-specific distribution pattern. The presence of cytokines in endometrial cells reinforces the view that these factors are involved in cell-cell interactions in endometrium. In addition, the temporal changes in the expression of cytokines suggests that the synthesis and/or secretion of these products may be under the regulatory influence of systemic steroid signals. Acknowledgements The authors thank Dr P.G.Salyaswaroop and S.Nicosia for reviewing this manuscript. This work is supported by Public Health Research Grant CA46866.
References Arend,W.P. (1991) Interleukin 1 receptor antagonist. A new member of the interleukin family. J. din. Invest., 88, 1445-1451. Arend.W.P., Welgus.H.G., Thompson.R.C. and Eisenberg.S.P. (1990) Biological properties of recombinant human monocyte-derived interleukin 1 receptor antagonist. J. Gin. Invest., 85, 1694—1697. Carter.D.B., Diebel,M.R.,Jr, Dunn.C.J., Tomich.C.-S.C, Laborde, A.L., Slightom,J.L., Berger.A.E., Bienkowski.M.J., Sun.F.F., McEwan.R.N., Harris,P.K.W., Yem.A.W., Waszak.G.A., OiosayJ.G., Sieu.L.C, Haidee.M.M., Zurher-Neety.H.A., Reankm, I.M., Heinrikson.R.L., Truesdell.S.E., Shelly J.A., Eessalu.T.E., Taylor.B.M. and Tracey.D.E. (1990) Purification, cloning, expression and biological characterization of an interleukin-1 receptor antagonist protein. Nature, 244, 633-638. Farrell.R.E.Jr (1989) Methodologies for RNA characterization. I: The isolation and characterization of mammalian RNA. Clin. Biotech., 1, 50-58. Fox.H.S., Bond.B.L. and ParslowJ.G. (1991) Estrogen regulates the IFN-7 promoter. J. Immunol., 146, 4362-4367. Hannum.C.H., Wicox.C.J., Arend,W.P., Joslin.F.G., Dripps.D.J.,
CytoUnes in endometrhmi Heimdal,P.L., Armes.L.G., Eisenberg.S.P. and Thompson,R-C. (1990) Interleukin-1 receptor antagonist activity of a human interleukin-1 inhibitor. Nature, 343, 336-340. Hsu.S.M., Raine,L. and Fanger,H. (1981) Use of avidiivbiotin-peroxidase complex (ABC) in immunoperoxidase techniques: A comparison between ABC and unlabeled antibody (PAP) procedures. J. Kstochem. Cytochem., 29, 577-580. HunU.S., Chen,H.-L., Hu,X.-L. and Tabibzadeh.S. (1992) Tumor necrosis factor-a mRNA and protein in human endometrium. BioL RepraL, 47, 141-147. Karyia.M., Kanzaki.H., Takakura.K., Imai.K., Okamoto.N., Emi.N., Kariya,Y. and Mori.T. (1991) Interieukin-1 inhibits in vitro AyTriimlrretion of human endometrial stromal cells. J. Gin. EndocrinoL Metab., 73, 1170-1174. Kauma.S., Dennis.M., Strom.S., Eierman.D. and Tumer.B.S. (1990) Interleukin-1/3, human leukocyte antigen HLA-DRa and transforming growth factor-/3 expression in endometrium, placenta and placenta! membranes. Am. J. Obstet. GynecoL, 163, 1430-1437. Lee.D.C. (1990) TGF-alpha. Expression and biological activities of the integral membrane precursor. MoL Reprod. Dev., 21, 37—45. Mclntyre.K.W., Stepan,G.J., Kolinsky.K.D., Benjamin.W.R., PlocinskiJ.M., Kafifka.K.L., Campen.C.A., Chizzonite.R.A. and Kilian.P.L. (1991) Inhibition of interleukin 1 (IL-1) binding and bioactivity in vitro and modulation of acute inflammation in vivo by IL-1 receptor antagonist and anti-IL-1 receptor monoclonal antibody. J. Exp. Med, 173, 931-939. Noyes.R.W. and Hertig^A.T. (1955) Dating the endometrial biopsy. FertiL
Steril., 1, 3-25. Pacifici.R., Rifes.L., McCracken,R., Vered.I., McMurtry.C, Avioh\L.V. and Peck.W.A. (1989) Ovarian steroid treatment blocks a postmenopausal increase in blood monocyte interleukin 1 release. Proc. Natl Acad Sci. USA, 86, 2398-2402. Pampfer.S., Tabibzadeh.S., Chuan,F.-C. and PoUardJ.W. (1991) Molecular cloning of a novel transcript for colony stimulating factor-1 from human endometrial glands. Production of a transmembrane form of the protein. J. MoL EndocrinoL, 5, 1931-1938. Ralston.S.H., Russell.R.G.G. and Gowen.M. (1990) Estrogen inhibits release of tumor necrosis factor from peripheral blood mononuclear cells in postmenopausal women. J. Bone Mineral Res., 5, 983-988. Rapopolee.D.A., Mark.D., Banda.MJ. and Werb.Z. (1988) Wound macrophages express TGF-a and other growth factors in vivo: Analysis by mRNA phenotyping. Science, 241, 708-712. Romero.R., Wu.K.Y., Brody.D.T., Oyarzun,E., Duff.G.W. and Drum.S.K. (1989a) Human decidua: A source of interleukin-1. Obstet. Gynecol., 73, 31-34. Romero.R., Ying King Wu, Oyarzun^E., HobbinsJ.C. and Mitchefl.M.D. (1989b) A potential role for epidermal growth factor/a-transforming growth factor in human parturition. Eur. J. Obstet. GynecoL Reprod. Biol., 33, 55-60. Romero.R., Manogue,K.R., Murray.D.M., Wu.K.Y., Oyarzun.E., Hobbins^.C. and Cerami,A. (1989c) Infection and labor. IV. Cachectintumor necrosis factor in the amniotic fluid of women with intraamniotic infection and preterm labor. Am. J. Obstet. Gynecol., 161, 336-341. Romero.R., Avila.C, Santhanam.U. and Sehgal.P.B. (1990) Amniotic fluid interleukin 6 in preterm labor. Association with infection. J. Qin. Invest., 85, 1392-1400. Saeki.T., CristanoA, Lynch.MJ., Brattain.M., Kim.N., Normarmo.N., Kenney.N., Ciardiello.F. and Salomon.D.S. (1992) Regulation by estrogen through the 5'-flanking region of the transforming growth factor a gene. MoL EndocrinoL, 5, 1955-1963. SambrookJ., Fritsch.E.F. and Maniatis.T. (1989) Molecular Cloning. A Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Tabibzadeh.S.S. (1990) Immunoreactivity of human endometrium: Correlation with endometrial dating. FertiL SteriL, 54, 624-631.
Tabibzadeh.S. (1991a) Human endometrium; an active site of cytoirine production and action. Endocr. Rev., 12, 272-290. Tabibzadeh.S. (1991b) Ubiquitous expression of TNF-a/Cachectin in human endometrium. Am. J. Reprod. ImmunoL, 26, 1 —4. Tabibzadeh.S.S. and Gerber.M.A. (1986) Immunohistologic analysis of lymphotd cells by a rapid double immuDoenzymatic labeling. J. Immunol. Methods, 91, 169-174. Tabibzadeh.S.S., Gerber.M.A. and Satyaswaroop^.G. (1986) Induction of HLA-DR antigen expression in human endometrial epithelial cells in vitro by recombinant gamma-interferon. Am. J. PathoL, 125, 9 0 - % . Tabibzadeh.S.S., Satyaswaroop,P.G. and Rao,P.N (1988) Antiproliferative effect of interferon gamma in human endometrial epithelial cells in vitro: Potential local growth modulatory role in endometrium. J. EndocrinoL Metab., 67, 131-138. Tabibzadeh.S.S., Santhanam.U., Sehgal,P.B. and May.L. (1989) Cytokine-induced production of interferon /32/interieukin-6 by freshly explanted human endometrial stromal cells. Modulation by estradiol-17/3. J. ImmunoL, 142, 3134-3139. Tabibzadeh.S.S., SivarajahA, Carpenter.D., Ohlsson-WnhelnvB.M. and Satyaswaroop^.G. (1990a) Modulation of HLA-DR expression in epithelial cells by interleukin 1 and estradiol-17/3. J. EndocrinoL Metab., 71, 740-747. Tabibzadeh.S.S., Kaffka.K.L., Satyaswaroop.P.G. and Kilian,P.L. (1990b) IL-1 regulation of human endometrial function: Presence of IL-1 receptor correlates with IL-1 stimulated PGE2 production. J.Endocrinol. Metab., 70, 1000-1006. Takacs.L., Kavacs.E.J., Smith.M.R., Young.H.A. and Durum,S.K. (1985) Detection of IL-la and IL-1/3 gene expression by in situ hybridization. Tissue localization of IL-1 mRNA in the normal C57BL/6 mouse. J. ImmunoL, 141, 3081-3095. Tamada.H., Das.S.K., Andrews.G.K. and Dey.S.K. (1991) Cell-type specific expression of transforming growth factor-a in the mouse uterus during the peri-implantation period. Biol. Reprod., 45, 365—372. Tamatani.T., Tsunoda.H., Iwasaki.H., Kaneko.M., Hashimoto.T. and Onozaki.K. (1988) Existence of both IL-la and /S in normal human amniotic fluid: Unique high molecular weight form of IL-1/3. Immunology, 65, 337-342. Wegmann.T.G., AthanassakisJ., Guflbert,L., Branch.D., Menu.M.D.E. and Chaouat.G. (1989) The role of m-CSF and GM-CSF in fostering placenta] growth, fetal growth and fetal survival. Transplant Proc., 21, 568-588. Received on April 16, 1992; accepted on June 16, 1992
1221
Cytokine expression in human endometrium throughout the menstrual cycle
S.Tabibzadeh1 and X.Z.Sun Department of Pathology, University of South Florida Health Sciences Center and Moffitt Cancer Center, Tampa, FL 33612, USA 'To whom correspondence should be addressed at: Department of Pathology, Moffitt Cancer Center, 12902 Magnolia Drive, Tampa, FL 33612, USA
Recent evidence suggests that diverse endometrial functions may be regulated by cytokines. In this report, the presence of protein and mRNA of cytokines were studied in human endometrium throughout the menstrual cycle. The presence of the interkukin-1 (1L-1)«, interieukin-1 (TL-l)jS, interieukin receptor antagonist (TRAP), interleukin-6 (IL-6) and transforming growth factor (TGF>a proteins were demonstrated by immunohistochemkal staining. The IL-la and TGF-a proteins were strongly expressed and IL-1/3 protein was weakly expressed in all the cells in the stroma as well as epithelial cells. IRAP was markedly expressed in the cells with morphological features of macrophages scattered in the stroma, and the expression of IL-6 protein was predominant in the endometrial epithelium. Diffuse cytoplasmic expression of IL-la in endometrial epithelium during the proliferative phase contrasted markedly with its enhanced luminal expression during the secretory phase of the menstrual cycle. In addition, the presence of the mRNA of these cytokines in endometrium was established throughout the entire menstrual cyde by reverse transcription-polymerase chain reaction (RT-PCR). Abundant expression of cytokines in human endometrium emphasizes the significant roles that cytokines play in cell-cell interactions and in regulating endometrial functions. Key words: cytokine/endometrium/growth polymerase chain reaction
factor/mRNA/
whether some of the effects of tiiese hormones are effects directed at the target tissue or are indirectly exerted through elaboration of other factors within the endometrium. It is conceivable that many cell-cell interactions and steroid-cell interactions in the endometrium occur via the production of factors that are collectively called cytokines. The emerging evidence suggests that many of the endometrial functions are controlled by these factors. Such diverse functions as proliferation, secretion of prostaglandin E2 (PGE2) or cytokines, human leukocyte antigen (HLA)-DR expression, and differentiation may be regulated by various cytokines (Tabibzadeh, 1991a; Tabibzadeh et al., 1986, 1988, 1990a,b; Karyia et al., 1991). In addition, these factors may be under the controlling influence of steroid hormones linking the effect of local factors to die systemic signals (Pacifici et al., 1989; Tabibzadeh et al., 1989; Ralston et al., 1990; Fox et al., 1991). In view of the overwhelming evidence for the contribution of the cytokines in the regulation of endometrial functions, this study was designed to demonstrate the expression of the proteins and the genes of several cytokines in human endometrium. Reverse transcription followed by polymerase chain reaction (RT-PCR) has emerged as an extremely sensitive and rapid technique which allows detection of low copy-number mRNA species including those of cytokines. In studies where the amount of available tissue for isolation of RNA is limited, this technique allows detection of the expression of multiple genes, a technique called mRNA phenotyping (Rapopolee et al., 1988). In view of our interest to show the presence of mRNA of several cytokines in small amounts of endometria, and particularly to verify their expression throughout the entire menstrual cycle, this technique seemed suitable. Subsequently, the expression of cytokine proteins was studied in situ by an immunohistochemical staining procedure. Materials and methods
Introduction Human endometrium is composed of glandular structures which are invested with endometrial stroma. Both components characteristically undergo a predictable series of structural, morphological, cytochemical and immunohistochemical changes (Tabibzadeh, 1990, 1991a). These changes are accompanied by sequences of proliferation (proliferative phase), secretion (secretory phase) and menstrual shedding (menstruation). It is thought that these changes are primarily driven by two steroid hormones, oestrogen and progesterone. However, it is unclear 1214
The mouse monoclonal antibodies to interieukin-la (IL-la), interieukin-1/3 (IL-1/3), and transforming growth factor-a (TGFa) were obtained from Olympus Immunochemicals (Lake Success, NY, USA). The monoclonal antibody to interieukin receptor antagonist protein (IRAP) was a generous gift of Dr M.J.Bienkowski (Upjohn Co., Kalamazoo, Michigan, USA). The IgG fraction of a polyclonal rabbit antiserum to interleukin-6 (IL-6) was kindly provided by Dr L.May (Rockefeller University, NY, USA). Primers to the IRAP were kindly provided by Dr D.B.Carter (Upjohn Co.). All other primers used in tius study were obtained from Clontech (Palo Alto, CA, USA). The © Oxford University Press
Cytokines in endometrium
TaWe I. Primers used in the polymerase chain reactions (PCR) and the size of the amplified products Name
Sequences of 5' and 3' primers
IL-la
5' ATGGCCAAAGTTCCAGACATGTTTG 3' GGTTTTCCAGTATCTGAAAGTCAGT
IL-10
IRAP
IL-6
TGF-a
0-Actin
Size of amplified products (bp)
5' ATGGCAGAAGTACCTAAGCTCGC 3' ACACAAATTGCATGGTGAAGTCAGTT
5' CCAGAAGACCTTCTATCTGAGGA 3' CCAGCTGGAGGCAGTTAAGATCAC
816 (Hireffll: 190;626) 802 (Hindm: 396;406) {TaqY. 314;488) (Nco\: 209;593) 201 (NcoY. 77; 124) (Sau96l: 53,148)
5' ATGAACTCCTTCTCCACAAGCGC 3' GAAGAGCCCTCAGGCTGGACTG (Taql: (XbaY.
628 182;446) 148;480)
(Rsal:
297 69;228)
5' ATGGTCCCCTCGGCTGGACAG 3' GGCCTGCTTCTTCTGGCTGGCA 5' ATGGATGATGATATCGCCGCGG 3' CTAGAAGCATTTGCGGTGGACGATGGAGGGGCC
1126
The lengths of the restriction enzyme-digested products (in bp) are indicated in parentheses. IL = interleukin; IRAP = interleuldn receptor antagonist protein; TGF = transforming growth factor.
sequences of the 5' and 3' primers and the lengths (in base pairs, bp) of the amplified products from RT-PCR reactions of cytokine mRNA are shown in Table I. The AMV reverse transcriptase was obtained from Boehringer Mannheim (Emeryville, CA, USA) and the Amplitaq was purchased from Cetus (Norwalk, CT, USA). Processing of endometria Endometria (n = 31) were from hysterectomy specimens from patients in their reproductive years. The uteri were removed for abnormalities which were not endometrial in origin, including uterine leiomyomas, and cervical or ovarian lesions. Upon removal, uteri were quickly transferred to the laboratory. Blocks of endometrium, 0.5 cm in diameter, were removed and embedded in paraffin. Sections were stained by haematoxylin and eosin and dated according to the published criteria (Noyes and Hertig, 1955). Identical blocks were placed in OCT compound, then snap frozen in a mixture of methylpropane and dry ice and kept at -70°C prior to immunostaining. Remaining endometrium was frozen and kept in liquid nitrogen for isolation of RNA. Endometria were dated with respect to early proliferative (n = 2), mid-proliferative (n = 3), late proliferative (n = 5), early secretory (n = 6), mid-secretory (n = 8) and late secretory (n = 7) phases of the menstrual cycle. Endometria used in this study showed normal morphology. Immunohistochemical staining procedure Air dried, 4-/tm-thick cryostat sections were immunostained by the avidin—biotin complex (ABC) staining method (Hsu et al., 1981; Tabibzadeh and Gerber, 1986). Briefly, the immunostaining procedure consisted of incubation with normal horse serum, mouse monoclonal or rabbit polyclonal antibodies, the appropriate biotinylated horse anti-mouse or biotinylated goat
anti-rabbit antibodies, ABC complexes and then development in a mixture of 4-chloro-l-naphthol (CN)-H2O2, resulting in a blue reaction product. The polyclonal antibody was used at 1/100 dilution and the monoclonal antibodies were used at 1 - 2 /*g/ml. Secondary antibodies were used at 1/200 dilution. Negative controls for immunostaining consisted of the substitution of primary antibody by irrelevant antibody, normal rabbit serum, rabbit or mouse immunoglobulins at equivalent protein concentrations or phosphate-buffered saline (PBS). Endogenous avidinbinding activity and endogenous peroxidase activity were demonstrated respectively by incubation with avidin-biotin complex and then development in a mixture of 4-chloro-lnaphthol (CNH^Qz or directly by incubation with the substrate. Alternatively, the endogenous peroxidase was quenched during the immunostaining procedure for 30 min in 0.3% H2C>2 in methanol following incubation of the section with the primary and then the biotinylated antibodies and prior to application of the ABC complexes. The tissues were examined without counterstaining. The intensity of the positive reaction with the immunostaining was semi-quantitatively expressed as: strong, moderate, weak and no staining. Isolation of RNA Total RNA was isolated by the guanidinium isothiocyanatecaesium chloride method from endometrial curettings weighing 20-100 mg (Farrell, 1989; Sambrook et al., 1989). Following digestion in DNase, a 15 /tg aliquot of the RNA was electrophoresed on formaldehyde denaturing gel to detect the intact 18S and 28S ribosomal RNA and to confirm the integrity of the endometrial RNA. The amount of RNA was calculated by measuring optical density at 260 nm in a spectrophotometer (Perkin-Elmer). 1215
S.Tabibzadeh and X.Z.Sun
Restriction enzyme digestion of amplified products
Table D. Cytoldne profiles in the endometria used in the reverse transcriptase—polymerase chain reaction (RT—PCR) IL-1 alpha
IRAP
IL-6
TGF-a
ND ND
ND ND
ND ND
ND ND
ND
ND
+ +-
Mid-P Mid-P Late-P Early-S Early-S Mid-S Mid-S Mid-S Mid-S Late-S Late-S
IL-1 beta
+
ND
ND
ND
+ +
Date of endometrium
ND
Abbreviations: P, proliferative; S, secretory; + , positive; ND, not done (insufficient amount of tissue); IL = interleukin; IRAP = interleultin receptor antagonist protein; TGF = transforming growth factor.
Reverse transcription Reverse transcription was performed as described (Rapopolee et al., 1988; Sambrook et al., 1989). Following heating at 65 °C for 5 min, 1 /ig aliquot of total RNA in 7.5 /J of diethylpyrocarbonate (DEPC)-treated water was used in each reverse transcription reaction. Reverse transcription was allowed in the presence of 5 units of AMV reverse transcriptase, 40 units of RNAsin, 5 /xl of 5 X reverse transcription buffer (500 mM Tris-HCl, pH 8.3, 250 mM KC1, 50 mM MgCl2, 50 mM dithiothreitol), 2.5 /J of random hexamer (0.5 /tg//J), 2.5 /d dNTP mixture (10 mM each). The total reaction volume was 17.5 id and the reaction was allowed to continue at 42°C for 1 h. Polymerase chain reaction The reverse transcription reaction mixture was used in a 100 /il PCR as follows. PCR was carried out by the addition of Amplitaq (2.5 IU), dNTP (200 /iM), primers (1 /*M), 10 /il of lOx reaction buffer (100 mM Tris-HCl, 500 mM KC1, 15 mM MgCl2, 0.01% gelatin) and water. The PCR reaction mixture was overlaid with mineral oil and amplification was carried out in a thermocycler (Perkin-Elmer/Cetus, Norwalk, CT, USA). The 30-cycle amplification profile consisted of denaturation at 95 °C for 1 min, annealing at 55°C for 2 min, extension at 72°C for 2 min with a final extension at 72°C for 5 min. The thermocycler was located distant from the preparation area and the amplified products were analysed in a separate laboratory. Fractions (10-20 /xl) of the amplified products along with *xl74 RF DNAJHaeUl fragments (1353, 1078, 872, 603, 310, 271; 281, 234, 194, 118, and 72 bp) serving as molecular weight standards were electrophoresed on a 2—2.5% Nusieve-agarose gel containing ethidium bromide and the bands were visualized and photographed in ultraviolet light. In each PCR, the RNA was amplified without the reverse transcription step to detect the presence of any contaminating genomic DNA. Also, a tube containing all the PCR components except the RT reaction mixture served as a negative control to check for the presence of DNA that may have been carried over from a prior reaction. Endometria which were used in the RT reactions are listed in Table n . 1216
The expected lengths (in bp) of the restriction enzyme-digested products of amplicons of cytokines were determined using a recombinant toolkit (Biosoft, Cambridge, UK). The band containing the amplified product from 200—300 /d PCR reactions was removed following electrophoresis and then extracted with phenol-chloroform and precipitated in a mixture of 100% ethanol and 1/10 volume of 3 mM sodium acetate. Following the enzyme digestions, at appropriate temperature for 30 min to 2 h, the samples were electrophoresed to determine the lengths of the digested products. The expected lengths (in bp) of the amplified products of the cytokines as well as their respective restriction enzyme-digested products are shown in Table I.
Results Cellular localization of cytoldne proteins The cytokine proteins could be localized in all the cases studied. Epithelial cells and all the cells in the stroma, including stromal, lymphoid and endothelial cells, expressed all these cytokines; however, the intensity of staining in cells of different lineages was different with respect to each specific cytokine. IL-1 a and TGF-a were strongly stained in equal intensity in cells of different lineages both in the epithelium and stroma and the endothelial cells (Figures 1-3, Table HI). The distribution pattern of immunostaining for IL-1/3 was similar to the staining profile of IL-la; however, the intensity of the immunostaining was weak to absent (Figure 4, Table IU). The strongest staining of IRAP was observed in cells with the morphological characteristics of macrophages, such as abundant cytoplasm and bean-shaped nuclei. These cells were scattered in the endometrial stroma (Figure 5, Table HI). The epithelial, stromal and endothelial cells exhibited weak to moderate staining for IRAP (Figure 5). The epithelial cells exhibited marked staining and the cells in the stroma showed weak to moderate staining for IL-6 (Figure 6, Table III). Some regional and gland—gland variations could be observed in the cases which were immunostained for cytokines. In control sections, no reaction product was present upon omission of the primary antibodies and their substitution with irrelevant antibodies, PBS or animal serum. However, endogenous peroxidase activity was observed in polymorphonuclear leukocytes and macrophages and pseudoperoxidase activity was present in the red blood cells in specifically stained sections as well as controls (Figure 6). Under the experimental conditions, no endogenous avidin-binding activity was observed in any of the cases studied.
Temporal changes in the expression of cytokine proteins Some menstrual cycle-related changes could be observed in the expression of the cytokine proteins. This was most evident in the expression of IL-la. The immunostaining for IL-la was rather uniformly expressed in the cytoplasm of the endometrial glands in the proliferative phase (Figure 1). In the mid- to late secretory phases, however, the immunoreactivity for IL-la shifted luminally (Figure 2).
Cytoldnes in endometriuro
FTg. 1. Immunoreactivity for interieukin-1 (IL-l)a is strongly expressed in the epithelium and the stroma in a late proliferan've endometrium (X200). Fig. 3. Immunoreactivity of the transforming growth factor (TGF-a) is strongly observed in cells in the stroma (small arrowheads), endothelial cells (large arrowheads) and epithelium (arrows) in this mid-proliferative endometrium (X160).
TaMe HI. Differential expression of cytokine proteins in epithelium versus stroma in human endometrium Intensity of staining
IL-la IL-1/3 IRAP TGF-a IL-6
Fig. 2. Immunoreactivity for interleukin-1 (IL-l)ar is observed in the luminal bonders of the epithelium in a mid-secretory endometrium (large arrowheads). Small arrowheads point to the positively immunostained cells in the stroma (xl60).
Epithelium
Stroma
Strong Weak Moderate Strong Strong
Strong Weak Weak (strong expression in macrophages) Strong Weak to moderate
IL = interieukin; IRAP = interieukin receptor antagonist protein; TGF = transforming growth factor.
i, .
Demonstration of cytotdne gene expression The IL-la, IL-l/S, IRAP, IL-6 and TGF-a gene expression was demonstrated by the reverse transcription of their respective mRNA to cDNA using random hexamers as primers for the AMV reverse transcriptase-mediated reaction followed by amplification using cytokine-specific primers (Table I). The presence of amplified products was detected by their mobility after electrophoresis in ethidium bromide-stained agarose gels and further identification of the amplified products was made by restriction enzyme digestion. Amplicons of all cytokines were observed in all the cases studied. The amplicons of amplified products from RT-PCR reactions for various cytokines were of the expected lengths (bp) (Table I, Figures 7-12). The lengths (in bp) of the restriction enzyme-digested products also conformed to those expected (Table I, Figures 7 — 12). Non-specific products were not observed in any of the RT—PCR reactions (Figures 7—12). The amplification of the /3-actin mRNA served as a positive control in all the RT-PCR reactions. In all reactions, and in all the cases studied, the amplified product for /3-actin
Fig. 4. Immunoreacu'vity for interleukin-1 (IL-l)/3 is weakly expressed in epithelium and the stroma in this proliferative endometrium (X100).
could be observed. Contaminating genomic DNA was not detected in PCR, as the total RNA amplified without the reverse transcription step did notrevealany amplified product (Figure 7). 1217
S.TaMbzaddi and X.Z.Sun
Fig. 5. The strongest expression of immunoreactive interleukin receptor antagonist protein (IRAP) is observed in some cells scattered in the stroma with morphological features of macrophages (large arrowheads). Epithelial cells (arrow) exhibit a moderate staining and cells in the stroma (small arrowhead) a weak staining in this early proliferative endometrium (X160).
Fig. 7. Electrophoresis of the products of the reverse transcription—polymerase chain reaction (RT—PCR) of interleukin (EL)-la, EL-1/3 and interleukin receptor antagonist protein (IRAP) mRNA of endometrium. Lane 1: amplification of mRNA without reverse transcription step. Lane 2: amplification product of EL-la mRNA (816 bp). Lane 3: amplification of mRNA without reverse transcription. Lane 4: amplified product of EL-1/3 mRNA (802 bp). Lane 5: xl74 replication factor (RF) DNAJHaeUl fragments. Lane 6: amplified product of IRAP mRNA (201 bp).
Fig. 6. Immunoreactive interleukin-(IL)-6 is strongly expressed in the glandular epithelial cells in this mid-secretory endometrium. The arrowhead points to an endogenous peroxidase positive cell (X160).
In addition, amplified product was not observed in PCR reactions where all the PCR components were used except for the addition of the RT reaction mixture.
Discussion In this report, the presence and distribution of cytokine-positive cells was shown by immunostaining of sections of endometria. The presence of the cytokine mRNAs was confirmed by RT—PCR and cells of different lineages expressed cytokine proteins. They included epithelial, endothelial, lymphoid and stromal cells. IL-la and TGF-a showed the widest distribution with respect to the cell types, the regional and temporal distributions. IL-1/3 showed a weak but similar distribution. This is in 1218
Fig. 8. Electrophoresis of the products of the reverse transcription—polymerase chain reaction (RT-PCR) of interleukin (IL)-la and IL-1/3 mRNA of endometrium. Lane 1: amplified product of IL-la mRNA (816 bp). Lane 2: amplified product of IL-la digested with MndHI (190 and 626 bp). Lane 3: amplified product of EL-1/3 mRNA (802 bp). Lane 4: The 396 and 406 bp products of Hindm digestion of the amplified product of IL-1/3 are overlapped. Lane 5: *xl74 RF DNA/WaeHI fragments.
Cytoklnes In endotnetrtuin
Fig. 9. Electrophoresis of the products of the reverse transcription—polymerase chain reaction (RT—PCR) of interieukin (IL)-/3 mRNA of endometrium. Lane 1: blank. Lane 2: amplified product of IL-1/9 mRNA (802 bp). Lane 3: amplified product of IL-l/S digested with Taql (314;488 bp). Lane 4: amplified product of IL-10 mRNA digested with Ncol (209;593 bp). Lane 5: xl74 RF DfiAJHaem fragments.
FTg. 11. Electrophoresis of the product of the reverse transcription-polymerase chain reaction (RT—PCR) of transforming growth factor (TGF)-a mRNA of endometrium. Lane 1: amplified product of TGF-a mRNA (297 bp). Lane 2: amplified product of TGF-a digested with Rsal (69 and 228 bp). Lane 3: *xl74 RF DNAJHaeUl fragments.
Fig. 10. Electrophoresis of the product of the reverse transcription—polymerase chain reaction (RT—PCR) of interieukin receptor antagonist protein (IRAP) mRNA of endometrium. Lane 1: amplified product of IRAP mRNA (201 bp). Lane 2: amplified product of IRAP digested with Ncol (77 and 124 bp). Lane 3: amplified product of IRAP mRNA digested with Sau96I (148;53 bp). Lane 4: *xl74 RF DNA//faem fragments.
F ^ . 12. Electrophoresis of the product of the reverse transcription—polymerase chain reaction (RT—PCR) of interieukin (IL)-6 mRNA of endometrium. Lane 1: *xl74 RF DtiAJHaeTa fragments. Lanes 2 and 5: amplified products of IL-6 mRNA (628 bp). Lane 3: amplified product of IL-6 digested with Xba\ (148 and 480 bp). Some undigested product has remained. Lane 4: blank. Lane 6: amplified product of IL-6 digested with Taql (182 and 446 bp).
line with the evidence that the mRNA of IL-1/3 is not observed by Northern blotting in the proliferative phase and is only rarely detected in the secretory phase in human endometria (Kauma et at., 1990). In contrast, IL-1/3 mRNA is easily detected in the gestational endometria and IL-1 activity may be detected in the human decidua and in the amniotic fluid (Romero et at., 1989a; Tamatani et at., 1988). In contrast to our finding of the wide distribution of IL-1 a and IL-1/3 in human endometrium, the message for IL-1 is apparently limited to cells with the
morphological characteristics of macrophages in mouse uteri (Takacs et at., 1985). Temporal and cell type-specific expression of the TGF-a transcript and protein have also been demonstrated in mouse uteri in the peri-implantation period by immunohistochemistry as well as by in-situ and Northern blot hybridizations (Lee, 1990; Tamada et at., 1991). This expression was shown both in the luminal and glandular epithelia on days 1 —4 and in many stromal cells on days 3 and 4 of pregnancy. The present 1219
S.TaWbzadeh and X.Z.Sun
study shows that in contrast to the equal intensity of immunoreactivity of IL-la and TGF-a in both epithelial and stromal cell components, IRAP is strongly expressed in cells in the stroma with morphological characteristics of macrophages whereas IL-6 is strongly expressed in the epithelial cells. Menstrual cycle-related changes in the expression of most cytokine proteins was not apparent This could be due to the semiquantitative nature of the immunostaining procedure. However, a menstrual cycle-related change in the expression of IL-la gene product was observed in the endometrial epithelium; the diffuse cytoplasmic staining in the proliferative phase shifted luminally in the secretory phase. This menstrual cycle-related change in the expression of a cytokine in human endometrium is not unprecedented. We have observed immunoreactivity for TNF-a in endometrial epithelial cells primarily in the secretory phase of the menstrual cycle. In addition, the immunoreactivity of TNF-a is strongest in stromal cells in the proliferative phase and is diminished in the secretory phase (Tabibzadeh, 1991b; Hunt et al, 1992). Also, the expression of the colony-stimulating factor (CSF)-1 transcripts in endometrial epithelium shows a distinct menstrual cycle-related pattern (Pampfer et al., 1991). Endometrium is extremely sensitive to steroid action and the sequential morphological, functional and immunochemical profiles of this tissue are primarily driven by the actions of steroids (Tabibzadeh, 1991a). Thus, it is conceivable that the menstrual cycle-related changes in the expression of cytokines in human endometrium also have a similar basis, hi accordance with this hypothesis is the demonstration of the blockage of release of both IL-1 and TNF-a by gonadal steroid treatment in monocytes (Pacifici et al., 1989; Ralston et al., 1990). Similarly, we have shown that the IL-1 mediated expression of IL-6 in human endometrial stromal cells may be down-regulated by oestrogen (Tabibzadeh et al., 1989). TGF-a transcript which is not detectable in non-pregnant mouse uteri, can be detected in oestrogen-treated uteri and high levels (50-70 ng/ml) of TGF-a protein are observed by Western blotting after diethylstilbestrol (DES) stimulation (Lee, 1990; Tamada et al, 1991). A recent study indicates that the 5'-flanking region of the TGF-a contains a potential oestrogen-responsive element(s) (Saeki et al., 1992). The ability of the oestradiol to regulate the interferon (IFN)gamma promoter and to increase the IFN-gamma mRNA expression in Con A-activated murine spleen cells provides further proof of interaction of steroid hormone with cytokines (Fox et al., 1991). Taken together, these data are in accordance with the hypothesis that steroid hormones may alter the expression of cytokines in endometrium and that some of the effects of oestrogen may be mediated through cytokines in this tissue. For example, it has been suggested that the mitogenic effect of oestradiol in endometrium may be mediated through epidermal growth factor (EGF)/TGF-a (Lee, 1990). With the recognition that many of the cell—cell interactions in endometrium may be driven by cytokines, diverse functions for the cytokines in endometrium are being described. These include control of growth, differentiation, secretion of prostaglandin-E2 and HLA-DR expression (Karyia et al., 1991; Tabibzadeh, 1991a; Tabibzadeh et al., 1986, 1988, 1989, 1990a,b). We have observed that the synthesis of IL-6 by the endometrial stromal cells may be induced by IL-1 as well as by 1220
IFN-gamma and TNF-a (Tabibzadeh et al., 1989). IL-1 also induces the synthesis of PGE2 in endometrial epithelial cells (Tabibzadeh et al., 1990b). Whereas progesterone drives the human endometrial stromal cells to become decidualized in vitro, IL-1 inhibits such differentiation (Karyia et al., 1991). IL-1 with its potent biological activity, is likely to be under stringent regulatory control in vivo. IRAP has now been recognized as one such regulatory mechanism (Carter et al., 1990; Hannum et al, 1990; Arend, 1991). IRAP blocks multiple biological activities of IL-1 including release of PGE2 by fibroblasts, synovia! cells and-chondrocytes, thymocyte proliferation, and collagenase production by synovial cells (Carter et al., 1990; Hannum etal., 1990; Arend et al., 1990; Arend, 1991; Mclntyre et al., 1991). Thus, IRAP may function in a similar manner in human endometrium to control the multiple actions of IL-1. The enhanced production of the placental IL-1 has been shown during labour and in intra-uterine infection and it has been suggested that this cytokine is involved in parturition (Romero et al., 1989a). It has also been speculated that TGF-a, IL-6, TNF-a, M-CSF (macrophage-colony stimulating factor) and GM-CSF (granulocyte-macrophage colony stimulating factor) are involved in the reproductive functions (Romero et al., 1989b, 1990; Tamada etal., 1991; Wegmann et al., 1989). In summary, we have demonstrated the expression of several cytokine genes and their respective products in human endometrium throughout the menstrual cycle. Cytokines were characterized by their cell-specific distribution pattern. The presence of cytokines in endometrial cells reinforces the view that these factors are involved in cell-cell interactions in endometrium. In addition, the temporal changes in the expression of cytokines suggests that the synthesis and/or secretion of these products may be under the regulatory influence of systemic steroid signals. Acknowledgements The authors thank Dr P.G.Salyaswaroop and S.Nicosia for reviewing this manuscript. This work is supported by Public Health Research Grant CA46866.
References Arend,W.P. (1991) Interleukin 1 receptor antagonist. A new member of the interleukin family. J. din. Invest., 88, 1445-1451. Arend.W.P., Welgus.H.G., Thompson.R.C. and Eisenberg.S.P. (1990) Biological properties of recombinant human monocyte-derived interleukin 1 receptor antagonist. J. Gin. Invest., 85, 1694—1697. Carter.D.B., Diebel,M.R.,Jr, Dunn.C.J., Tomich.C.-S.C, Laborde, A.L., Slightom,J.L., Berger.A.E., Bienkowski.M.J., Sun.F.F., McEwan.R.N., Harris,P.K.W., Yem.A.W., Waszak.G.A., OiosayJ.G., Sieu.L.C, Haidee.M.M., Zurher-Neety.H.A., Reankm, I.M., Heinrikson.R.L., Truesdell.S.E., Shelly J.A., Eessalu.T.E., Taylor.B.M. and Tracey.D.E. (1990) Purification, cloning, expression and biological characterization of an interleukin-1 receptor antagonist protein. Nature, 244, 633-638. Farrell.R.E.Jr (1989) Methodologies for RNA characterization. I: The isolation and characterization of mammalian RNA. Clin. Biotech., 1, 50-58. Fox.H.S., Bond.B.L. and ParslowJ.G. (1991) Estrogen regulates the IFN-7 promoter. J. Immunol., 146, 4362-4367. Hannum.C.H., Wicox.C.J., Arend,W.P., Joslin.F.G., Dripps.D.J.,
CytoUnes in endometrhmi Heimdal,P.L., Armes.L.G., Eisenberg.S.P. and Thompson,R-C. (1990) Interleukin-1 receptor antagonist activity of a human interleukin-1 inhibitor. Nature, 343, 336-340. Hsu.S.M., Raine,L. and Fanger,H. (1981) Use of avidiivbiotin-peroxidase complex (ABC) in immunoperoxidase techniques: A comparison between ABC and unlabeled antibody (PAP) procedures. J. Kstochem. Cytochem., 29, 577-580. HunU.S., Chen,H.-L., Hu,X.-L. and Tabibzadeh.S. (1992) Tumor necrosis factor-a mRNA and protein in human endometrium. BioL RepraL, 47, 141-147. Karyia.M., Kanzaki.H., Takakura.K., Imai.K., Okamoto.N., Emi.N., Kariya,Y. and Mori.T. (1991) Interieukin-1 inhibits in vitro AyTriimlrretion of human endometrial stromal cells. J. Gin. EndocrinoL Metab., 73, 1170-1174. Kauma.S., Dennis.M., Strom.S., Eierman.D. and Tumer.B.S. (1990) Interleukin-1/3, human leukocyte antigen HLA-DRa and transforming growth factor-/3 expression in endometrium, placenta and placenta! membranes. Am. J. Obstet. GynecoL, 163, 1430-1437. Lee.D.C. (1990) TGF-alpha. Expression and biological activities of the integral membrane precursor. MoL Reprod. Dev., 21, 37—45. Mclntyre.K.W., Stepan,G.J., Kolinsky.K.D., Benjamin.W.R., PlocinskiJ.M., Kafifka.K.L., Campen.C.A., Chizzonite.R.A. and Kilian.P.L. (1991) Inhibition of interleukin 1 (IL-1) binding and bioactivity in vitro and modulation of acute inflammation in vivo by IL-1 receptor antagonist and anti-IL-1 receptor monoclonal antibody. J. Exp. Med, 173, 931-939. Noyes.R.W. and Hertig^A.T. (1955) Dating the endometrial biopsy. FertiL
Steril., 1, 3-25. Pacifici.R., Rifes.L., McCracken,R., Vered.I., McMurtry.C, Avioh\L.V. and Peck.W.A. (1989) Ovarian steroid treatment blocks a postmenopausal increase in blood monocyte interleukin 1 release. Proc. Natl Acad Sci. USA, 86, 2398-2402. Pampfer.S., Tabibzadeh.S., Chuan,F.-C. and PoUardJ.W. (1991) Molecular cloning of a novel transcript for colony stimulating factor-1 from human endometrial glands. Production of a transmembrane form of the protein. J. MoL EndocrinoL, 5, 1931-1938. Ralston.S.H., Russell.R.G.G. and Gowen.M. (1990) Estrogen inhibits release of tumor necrosis factor from peripheral blood mononuclear cells in postmenopausal women. J. Bone Mineral Res., 5, 983-988. Rapopolee.D.A., Mark.D., Banda.MJ. and Werb.Z. (1988) Wound macrophages express TGF-a and other growth factors in vivo: Analysis by mRNA phenotyping. Science, 241, 708-712. Romero.R., Wu.K.Y., Brody.D.T., Oyarzun,E., Duff.G.W. and Drum.S.K. (1989a) Human decidua: A source of interleukin-1. Obstet. Gynecol., 73, 31-34. Romero.R., Ying King Wu, Oyarzun^E., HobbinsJ.C. and Mitchefl.M.D. (1989b) A potential role for epidermal growth factor/a-transforming growth factor in human parturition. Eur. J. Obstet. GynecoL Reprod. Biol., 33, 55-60. Romero.R., Manogue,K.R., Murray.D.M., Wu.K.Y., Oyarzun.E., Hobbins^.C. and Cerami,A. (1989c) Infection and labor. IV. Cachectintumor necrosis factor in the amniotic fluid of women with intraamniotic infection and preterm labor. Am. J. Obstet. Gynecol., 161, 336-341. Romero.R., Avila.C, Santhanam.U. and Sehgal.P.B. (1990) Amniotic fluid interleukin 6 in preterm labor. Association with infection. J. Qin. Invest., 85, 1392-1400. Saeki.T., CristanoA, Lynch.MJ., Brattain.M., Kim.N., Normarmo.N., Kenney.N., Ciardiello.F. and Salomon.D.S. (1992) Regulation by estrogen through the 5'-flanking region of the transforming growth factor a gene. MoL EndocrinoL, 5, 1955-1963. SambrookJ., Fritsch.E.F. and Maniatis.T. (1989) Molecular Cloning. A Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Tabibzadeh.S.S. (1990) Immunoreactivity of human endometrium: Correlation with endometrial dating. FertiL SteriL, 54, 624-631.
Tabibzadeh.S. (1991a) Human endometrium; an active site of cytoirine production and action. Endocr. Rev., 12, 272-290. Tabibzadeh.S. (1991b) Ubiquitous expression of TNF-a/Cachectin in human endometrium. Am. J. Reprod. ImmunoL, 26, 1 —4. Tabibzadeh.S.S. and Gerber.M.A. (1986) Immunohistologic analysis of lymphotd cells by a rapid double immuDoenzymatic labeling. J. Immunol. Methods, 91, 169-174. Tabibzadeh.S.S., Gerber.M.A. and Satyaswaroop^.G. (1986) Induction of HLA-DR antigen expression in human endometrial epithelial cells in vitro by recombinant gamma-interferon. Am. J. PathoL, 125, 9 0 - % . Tabibzadeh.S.S., Satyaswaroop,P.G. and Rao,P.N (1988) Antiproliferative effect of interferon gamma in human endometrial epithelial cells in vitro: Potential local growth modulatory role in endometrium. J. EndocrinoL Metab., 67, 131-138. Tabibzadeh.S.S., Santhanam.U., Sehgal,P.B. and May.L. (1989) Cytokine-induced production of interferon /32/interieukin-6 by freshly explanted human endometrial stromal cells. Modulation by estradiol-17/3. J. ImmunoL, 142, 3134-3139. Tabibzadeh.S.S., SivarajahA, Carpenter.D., Ohlsson-WnhelnvB.M. and Satyaswaroop^.G. (1990a) Modulation of HLA-DR expression in epithelial cells by interleukin 1 and estradiol-17/3. J. EndocrinoL Metab., 71, 740-747. Tabibzadeh.S.S., Kaffka.K.L., Satyaswaroop.P.G. and Kilian,P.L. (1990b) IL-1 regulation of human endometrial function: Presence of IL-1 receptor correlates with IL-1 stimulated PGE2 production. J.Endocrinol. Metab., 70, 1000-1006. Takacs.L., Kavacs.E.J., Smith.M.R., Young.H.A. and Durum,S.K. (1985) Detection of IL-la and IL-1/3 gene expression by in situ hybridization. Tissue localization of IL-1 mRNA in the normal C57BL/6 mouse. J. ImmunoL, 141, 3081-3095. Tamada.H., Das.S.K., Andrews.G.K. and Dey.S.K. (1991) Cell-type specific expression of transforming growth factor-a in the mouse uterus during the peri-implantation period. Biol. Reprod., 45, 365—372. Tamatani.T., Tsunoda.H., Iwasaki.H., Kaneko.M., Hashimoto.T. and Onozaki.K. (1988) Existence of both IL-la and /S in normal human amniotic fluid: Unique high molecular weight form of IL-1/3. Immunology, 65, 337-342. Wegmann.T.G., AthanassakisJ., Guflbert,L., Branch.D., Menu.M.D.E. and Chaouat.G. (1989) The role of m-CSF and GM-CSF in fostering placenta] growth, fetal growth and fetal survival. Transplant Proc., 21, 568-588. Received on April 16, 1992; accepted on June 16, 1992
1221
