Abstract
One of the most prominent inflammatory reactions is the activation of the complement system. The activated complement system does not distinguish between pathogens and the host cell. In order to prevent autologous complement-mediated attack, host cells express a variety of both membrane-bound and fluid-phase complement regulatory proteins which control activity of the complement cascade by acting on convertase enzymes or the membrane-attack complex. Although the process of ovulation is facilitated by the inflammatory reaction, this reaction has the potential to cause serious damage to growing follicles, ovulated follicles, and other important ovarian tissues. This study was undertaken to characterize the expression and regulation of decay-accelerating factor (DAF), a complement regulator, as a potential mediator of ovarian tissue protection from ovulatory inflammation. DNA microarray and Northern blot analyses showed that an ovulatory gonadotropin stimulus dramatically yet transiently induced DAF mRNA expression in the immature rat ovary. Northern blot and PCR analyses revealed that of the three known DAF isoforms, glycosylphosphatidylinositol (GPI)-, soluble-, and transmembrane-(TM) DAF, GPI-DAF was the predominant form. In situ hybridization localized GPI-DAF mRNA expression in the theca-interstitial cells of the periovulatory ovary. Neither the anti-progestin RU486 nor the cyclooxygenase inhibitor indomethacin significantly inhibited human chorionic gonadotropin (hCG)-induced GPI-DAF mRNA expression in vivo. In vitro theca cell culture studies indicated that hCG induces GPI-DAF mRNA expression through the protein kinase A pathway. This study suggests that gonadotropin-induced GPI-DAF may be involved in the protection of ovarian tissues from the potential attack by the complement system activated by the inflammatory response associated with ovulation.
Introduction
Complement, a central element of innate immunity, initiates and coordinates immediate immune reactions which protect the body from microbes, foreign particles, and altered self cells. It consists of a series of plasma proteins that when activated directly or indirectly destroy invading organisms and mediate humoral and cellular interactions of the immune response, including chemotaxis, phagocytosis, cell adhesion, and B cell differentiation (Walport 2001). The complement system can be activated via three major pathways: the classical pathway, the alternative pathway, and the lectin pathway. The classical pathway is stimulated by antibody–antigen complexes, the alternative pathway by spontaneous hydrolysis of native complement factor C3, and the lectin pathway by mannose-binding lectins which recognize microbial saccharides (Nauta et al. 2004). Although the preliminary steps of each pathway differ, all three generate a C3 convertase enzyme, which cleaves C3 into C3a and C3b. The C3b fragment generates the C5 convertase enzyme, allowing subsequent formation of the membrane-attack complex (MAC, C5b-9), which inserts itself into cell membranes causing damage by cytolytic mechanisms (Cole & Morgan 2003). Products produced during the complement cascade, such as the anaphylatoxins C5a and C3a, are capable of binding to receptors on mast cells, inducing them to release histamines and prostaglandins (PGs), and can serve as chemoattractants for neutrophils, eosinophils, and leukocytes (Kuby 1997, Nauta et al. 2004). Besides causing cell lysis, C5b-9 can promote production and release of cytokines, PGs, leukotrienes, and reactive oxygen species (Nauta et al. 2004).
As the complement system is non-selective, unable to distinguish between pathogen and host cell, host tissues have constructed a defense mechanism consisting of both membrane-bound and fluid-phase complement regulatory proteins, which check the activity of the complement cascade at the level of convertase enzymes and MAC (Liszewski et al. 1996, Hourcade et al. 2000). Complement regulators include membrane cofactor protein (CD46), membrane inhibitor of reactive lysis (CD59), and decay-accelerating factor (DAF, CD55). DAF and CD59 are detected in various tissues, while CD46 is detected predominantly in testes (Kumar et al. 1993, Mead et al. 1999). In the human ovary, CD46 and DAF protein are found in granulosa cells of primordial follicles, and in both granulosa and theca cells of developing follicles, while CD59 is only localized to granulosa and theca cells of developing follicles (Oglesby et al. 1996). In the rat, DAF protein was reported to be present in the endothelium of the ovary (Spiller et al. 1999). DAF provides the first line of defense against complement-mediated damage to host cells by accelerating the decay of both the classical and alternative pathway C3 and C5 convertases. In primates and rodents, several protein isoforms of DAF have been identified. The predominant form of the protein is the glycosylphosphatidylinositol-anchored form (GPI-DAF), but alternative splicing produces a transmembrane-domain containing form (TM-DAF) and a secreted form (soluble-DAF) (Hinchliffe et al. 1998, Miwa et al. 2000). When the complement attack overcomes this DAF-mediated defense, it is vital that cells possess a second line of defense in order to minimize the damaging effects of the MAC. CD59, another GPI-anchored protein, prevents MAC assembly by binding to component C8 and/or C9, preventing lytic pore formation (Miwa & Song 2001).
Inflammation has been postulated to be a key component of ovulation (Espey 1980, Ness et al. 2000), whereby the oocyte is released from the interior of the follicle (Brannstrom & Enskog 2002). For example, there is a rapid increase in the leukocyte population (Brannstrom & Enskog 2002), upregulation of pro-inflammatory gene expression (Richards et al. 2002), an increase in inflammatory cytokines (Norman & Brannstrom 1996), and increased proteolytic activities in the periovulatory ovary (Curry & Osteen 2003), all of which have been regarded as an indication of the existence of an acute inflammatory reaction at the time of ovulation (Espey 1980). In addition, numerous reports have shown that treatment with anti-inflammatory agents decreases the ovulation rate (Brannstrom & Enskog 2002), further supporting the idea that inflammation is important for successful ovulation. Complement components have been documented in the reproductive tract during the menstrual cycle. Indeed, there is evidence that active complement is necessary for ovulation to proceed normally. Females suffering from the autosomal dominant disorder hereditary angioedema, in which complement function deteriorates, often display polycystic or multifollicular ovaries (Perricone et al. 1992). While the inflammation associated with ovulation facilitates follicular rupture, it could also damage nearby ovarian tissue by its own proteolytic and cytolytic nature. Thus, we hypothesized that complement regulators should be upregulated during the periovulatory period in order to protect the ovarian tissues from attack of activated complement at the time of ovulation. In this study, we have characterized the ovarian expression of a critical component of this protective system, namely DAF, in association with factors involved in ovulation.
Materials and Methods
Materials
Pregnant mare’s serum gonadotropin (PMSG), human chorionic gonadotropin (hCG), RU486, and indomethacin were purchased from Sigma. Cell strainers of 40 and 70 μm were supplied by Becton Dickinson Falcon (Billerica, MA, USA). Restriction enzymes, RNaseOUT, and MLV reverse transcriptase were purchased from Invitrogen. [α-35S]UTP was provided by Milan Panic Biomedicals (Irvine, CA, USA). [α-32P]UTP was purchased from Amersham. Media for cell culture was obtained from Gibco.
Animals and treatments
Animal procedures were carried out in accordance with the University of Kentucky Animal Care and Use Committee. Twenty-day-old female Sprague–Dawley rats were purchased from Harlan Sprague–Dawley, Inc. (Harlan, IN, USA). The animals were kept in a 14 h light:10 h darkness cycle and given water and rat chow. Using the gonadotropin-primed superovulation model, 21-day-old rats were injected with 10 IU PMSG, then 48 h later, injected with 10 IU hCG. For DNA micro-array analysis in construction of the Rat Ovarian Gene Expression Database (rOGED, see below), tissue samples were collected at PMSG 0, 12 and 48 h and hCG 6 and 12 h; five animals were used per time point. For total RNA extraction, one ovary from each of four animals was collected at each time point. The remaining six ovaries were used for granulosa cell isolation and residual ovarian cell collection (Ko et al. 1999). As it is nearly impossible to isolate exclusively pure populations of a specific ovarian cell type in a short amount of time, there may exist minor crossover of cell types in the granulosa or residual ovarian cell samples. Uteri and oviducts from both ovaries of each animal were also collected and used for microarray analysis. For in situ hybridization, ovaries with intact oviducts were harvested at PMSG 48 h and hCG 6, 12 and 24 h, then stored at −80 °C for later sectioning. In order to study the effect of progesterone (P4) and PGs on GPI-DAF expression, animals were primed with PMSG, followed by injection of RU486 (2 mg/kg body weight) at 47 h post-PMSG and injection of hCG at 48 h post-PMSG. Indomethacin (1 mg/kg body weight) was administered 3 h after hCG injection. Rats were killed at PMSG 48 h, hCG 6 and 24 h. Oviducts of rats killed at hCG 24 h were examined and the number of oocytes counted to validate efficacy of treatments. One animal was used for each time point (n=3). One ovary from hCG 6 h, hCG 6 h+RU486 and hCG 6 h+indomethacin was also frozen at −80 °C and later used for in situ hybridization.
rOGED
The expression profile of GPI-DAF was identified using rOGED (http://web5.mccs.uky.edu/kolab/rogedendo.aspx), which was previously developed in our laboratory. The procedures used are described in detail elsewhere (Jo et al. 2004). The database provides the gene expression profiles of all genes and expressed sequence tags included in the Affymetrix Rat Expression 230A and 230B Gene-Chips. The microarray data is displayed in graphical format, showing the expression pattern over a time course of PMSG 0, 12 and 48 h and hCG 6 and 12 h for whole ovary, granulosa cells, and residual (non-granulosa) cells.
Northern analysis
Northern analysis was performed to determine the temporal expression of the GPI-DAF transcript, the isoform-specific expression of DAF mRNA, and the possible regulation of GPI-DAF by the P4 and PG pathway. Total RNA was extracted from the whole ovary, granulosa cells, or residual tissue using Trizol reagent according to the manufacturer’s protocol, then quantified by spectrophotometry. Five to ten micrograms of total RNA were separated on a 1% MOPS-formaldehyde agarose gel. The RNA was transferred to a nylon membrane (Schleicher & Schuell Inc., Keene, NH, USA) and fixed to the membrane by UV cross-linking (UV crosslinker, UVP CL-1000; VWR, West Chester, PA, USA). The blot was briefly washed in water, then air dried to remove residual 20 × SSC. Plasmids containing partial cDNA for the gene of interest were linearized with appropriate restriction enzymes. Antisense RNA probe was made by in vitro transcription using SP6 or T7 polymerase (Invitrogen) and [α-32P]UTP (10 mCi/ml). The probe was purified with mini Quick Spin RNA columns (Roche). The membrane was prehybridized in UltraHyb hybridization buffer (Ambion Inc., Austin, TX, USA) for 30 min at 68 °C, then probe was added to a concentration of 106 c.p.m./ml and hybridized at 68 °C for 12–16 h. Blots were washed twice (5 min each) with shaking in 2 × SSC, 0.1% SDS at room temperature, then washed at high stringency (0.1 × SSC, 0.1% SDS) during two 15 min washes at 68 °C. The membrane was exposed to film (Kodak Biomax XAR) for 48 h at −80 °C.
Theca cell isolation and culture
Ovaries were harvested from immature rats primed with PMSG for 48 h. Granulosa cells were removed from the ovaries as described previously (Ko et al. 1999). Briefly, the ovaries were incubated in preincubation media for 20 min at 37 °C to allow dissociation of granulosa cells, washed with cold serum-free 4F media (15 mM Hepes (pH 7.4), 50% DMEM and 50% Ham’s F12 with bovine transferrin (5 μg/ml), human insulin (2 mg/ml), hydrocortisone (40 ng/ml) and antibiotics) three times, and granulosa cells were removed by follicular puncture. Theca cells were then isolated from the residual ovarian tissue by the discontinuous density gradient centrifugation method as described by Magoffin & Erickson (1988). Cells were plated at a concentration of 200 000/well in McCoy’s media containing 1 × penicillin–streptomycin and 1× l-glutamine and incubated in a humidified atmosphere of 5% CO2 at 37 °C. Cells were treated immediately after plating to achieve optimal luteinizing hormone (LH) receptor (LH-R) responsiveness. Treatments were as follows: hCG (0.1 IU/ml), cycloheximide (CHX, 10 μg/ml), forskolin (protein kinase A (PKA) pathway activator, 10 μM) and phorbol 12-myristate 13-acetate (PMA, PKC pathway activator, 20 nM). Cells were incubated for 6 h after treatment, then harvested and RNA isolated using Trizol reagent. The effectiveness of 10 μg/ml CHX in blocking protein translation was demonstrated with a [35S]methionine incorporation assay. Briefly, cells were treated with CHX or vehicle (ethanol), then cultured with media containing [35S]methionine for 6 h. Cells were washed with PBS, precipitated with trichloroacetic acid, and activity measured in a scintillation counter.
RT-PCR
For the study of regulatory factors in GPI-DAF expression, RT-PCR was performed on total RNA extracted from theca cell cultures by the Trizol method. Total RNA (0.5–1 μg) was reverse transcribed using 250 ng random hexamer and 200 U MLV reverse transcriptase as routinely performed in our laboratory (Ko et al. 2003). PCR amplification was done at 20 and 25 cycles on an Eppendorf Mastercycler. The PCR products were separated on a 1.5% agarose gel, stained with SYBR Green I (Sigma) for 20 min, and scanned on a phosphorimager (FujiFilm FLA-5000). The identity of the RT-PCR bands was confirmed by Southern blotting using [α-32P]UTP-labeled GPI-DAF riboprobe. The detailed procedure is described elsewhere (Ko & Park-Sarge 2000). Primers for GPI-DAF were 5′-cta atg cca cgc caa act cg-3′ and 5′-cag ctt gta ccc ttt gtc gc-3′, numbered as 131–150 and 572–591 in the sequence AB026903; for TM-DAF were 5′-gcc tta ggg act act ata gg-3′ and 5′-gcc gtc atc taa ttc aca gg-3′, numbered as 2557–2576 and 2830–2849 in the sequence AB032395; for soluble-DAF were 5′-gcc aat cag tca ggt agc ac-3′ and 5′-ccc tta cca ttt cgt tca gg-3′, numbered as 1350–1369 and 1711–1730 in the sequence AF039584; and for ribosomal protein L19 were 5′-ggc tac aga aga ggc ttg cc-3′ and 5′-cat atg cct gcc ctt ccg-3′. Primers were synthesized by IDT (Integrated DNA Technologies, Coralville, IA, USA).
In situ hybridization
Frozen ovaries were sectioned at 10 μm on a Bright OTF cryostat and mounted onto Superfrost/Plus Microscope slides (VWR, West Chester, PA, USA). Sections were fixed, pretreated, and hybridized with antisense RNA probe as described previously (Ko et al. 2003). Plasmids containing cDNA for the gene of interest in the pCRII-TOPO vector (Invitrogen) were linearized with the appropriate restriction enzyme, then [α-35S]UTP-labeled RNA probes were synthesized using T7 or Sp6 polymerase. RNA probe (2 × 107 c.p.m./ml) in hybridization buffer was applied to sections and incubated at 42 °C in a humidity chamber for 15–18 h. Slides were washed and dehydrated as previously described (Ko et al. 2003). The slides were exposed to Kodak Biomax XAR film for 4 days and dipped in Kodak NTB-2 emulsion (Eastman Kodak, Rochester, NY, USA) for autoradiography. The slides were exposed for 4–6 weeks at 4 °C. After developing with Kodak D19, slides were stained with Gill’s No. 1 hematoxylin (Electron Microscopy Sciences, Fort Washington, PA, USA) and 1 mg/ml eosin (Sigma). The signal was visualized on an Olympus CKX41 microscope.
Statistics
For the in vitro theca cell culture experiment, the effect of treatments on GPI-DAF expression was analyzed by one-way ANOVA or Student’s t-test. GPI-DAF expression in non-treated, hCG- and hCG+CHX-treated cells or non-treated, forskolin- and forskolin+CHX-treated cells was assessed by one-way ANOVA using the University of Kentucky StatServer (http://sta.kbrin.uky.edu). The Student’s t-test was performed for the time course GPI-DAF expression using Microsoft Excel. P values of less than 0.05 were regarded as significant.
Results
Identification of gonadotropin-stimulated DAF mRNA expression in the periovulatory ovary
Recently, we have generated rOGED, which provides genome-wide temporal mRNA expression profiles in the intact ovary, granulosa cell, and residual ovarian tissue (ovarian cell types excluding granulosa cell) simultaneously (Jo et al. 2004) (http://web5.mccs.uky.edu/kolab/rogedendo.aspx). The rOGED profile revealed that hCG dramatically, yet transiently, induced GPI-DAF mRNA expression in the ovary (Fig. 1A). GPI-DAF mRNA expression was increased 6-fold by 6 h post-hCG administration in the residual tissue, while the expression was barely detectable in the granulosa cells. Northern blot analysis showed a spatio-temporal GPI-DAF mRNA expression pattern identical to that of the rOGED profile (Fig. 1B).
Isoforms of DAF mRNA
The Northern analysis detected 4.3, 2.9, and 1.6 kb transcript bands (Fig. 1B). To determine whether each of the three transcripts represents a different isoform of DAF mRNAs, three identical RNA blots were hybridized with GPI-, TM-, and soluble-form specific probes respectively. The GPI-form specific probe detected all three of the bands, while no band was detected by the TM- or soluble-form specific probe (Fig. 2). This result was confirmed by a subsequent RT-PCR assay, in which simultaneous amplification of GPI-DAF and TM-DAF or GPI-DAF and soluble-DAF in the same PCR tube resulted in predominant amplification of GPI-DAF over TM-DAF (Fig. 3) and soluble-DAF (not shown).
Spatio-temporal DAF mRNA expression in the ovary and oviduct
To localize the in vivo DAF mRNA expression, in situ hybridization was performed using tissue sections of gonadotropin-primed immature rat ovaries with attached oviducts. The strongest GPI-DAF mRNA signal was detected in the theca-interstitial cells of ovaries collected 6 h after hCG (Fig. 4). As was predicted from the Northern blot analysis, mRNA signal was detected only by the GPI-DAF probe (Fig. 4); no detected signal was observed in the tissue sections hybridized with TM-and soluble-DAF probes (not shown). The expression decreased to basal level by 24 h after hCG (Fig. 4A). No prominent GPI-DAF mRNA signal was seen in the granulosa cell layer (Fig. 4B); however, strong GPI-DAF mRNA signal was detected in the luminal epithelium of the oviduct (Fig. 4C).
P4 and PG do not induce DAF expression
PG and P4 are critically involved in ovulation (Richards et al. 1998). To test whether these hormones have a regulatory role in GPI-DAF expression in the ovary, we measured GPI-DAF mRNA expression levels in the RU486- or indomethacin-treated immature rat ovary. Ovaries were collected at 6 and 24 h post-hCG administration and were examined for GPI-DAF mRNA expression (Fig. 5). The effect of RU486 and indomethacin treatment on the ovulatory process was validated by checking for the presence of oocytes in the oviduct of 24 h post-hCG animals. Twenty-five to thirty oocytes were retrieved from each oviduct of the control group (hCG only), while approximately half that number were seen in oviducts of indomethacin-treated rats, and no oocytes were seen in those of RU486-treated animals. Northern blot (Fig. 5B) and in situ hybridization (Fig. 5C) analyses showed that there was no marked reduction of GPI-DAF mRNA expression in either RU486- or indomethacin-treated animals.
Regulatory factors on DAF expression
To determine which signaling pathways are involved in the hCG-induced GPI-DAF mRNA expression in theca cells, we analyzed thecal cell responsiveness to hCG treatment in vitro using cells isolated from PMSG 48 h primed immature rats. First, we measured the GPI-DAF mRNA transcript in theca cells which had not been treated over a time period of 72 h. Immediately after isolation, the theca cells displayed a basal level of GPI-DAF mRNA expression. Interestingly, however, significant induction of GPI-DAF mRNA was detected within 6 h of culture, after which it decreased considerably but increased again by 24 h (Fig. 6A). Due to the lengthy cell isolation procedure (4 h), and the fact that the LH-R expression peaks at PMSG 48 h, cells were treated immediately after plating to avoid losing LH-R activity. Although the expression level was slightly increased by the hCG treatment (Fig. 6B), the level of increase was substantially lower than was observed in vivo. When cultured in the presence of the PKA pathway activator forskolin or the PKC activator PMA for 6 h, no significant increase was observed (Fig. 6B). To ascertain whether or not intermediary protein synthesis was necessary for hCG or the PKA pathway to upregulate DAF expression, cells were treated with hCG or forskolin in the presence or absence of CHX, a protein translation inhibitor. Surprisingly, addition of CHX to either forskolin or hCG treatment significantly increased the GPI-DAF mRNA expression, while treatment with hCG or forskolin alone did not show a significant increase (Fig. 6B).
CD59 is expressed in the periovulatory ovary
We wondered if other complement regulators showed the same spatio-temporal expression as GPI-DAF. In the ovary, the complement regulator CD59 shows a different expression profile compared with GPI-DAF; while highly expressed in residual tissues from PMSG 0 h to hCG 12 h, it sharply increases in granulosa cells after hCG 6 h (Fig. 7).
Discussion
In this study, we found that the mRNA transcript for GPI-DAF increases dramatically by 6 h following hCG treatment, and this increase in expression occurs specifically in the theca-interstitial layer of the rat ovary. The upregulation of this complement regulator may be significant as a variety of inflammatory mediators and complement factors have been detected in the periovulatory ovary, many specifically in the theca layer. In the rat, neutrophils increase 8-fold in the theca layer of ovulating follicles, while macrophages increase 5-fold (Brannstrom & Enskog 2002). The expression of several chemokines and cytokines increases in the rat ovary at hCG 6 h, including monocyte chemotactic protein-1 and -3, macrophage inflammatory proteins-1α, -1β, and –1γ (Wong et al. 2002), and the interleukin-8 (IL-8)-like neutrophil chemoattractant CINC/gro (Ushigoe et al. 2000). Protein expression of both CINC/gro and its human counterpart IL-8 was localized to the theca layer (Runesson et al. 2000). Functionally active complement is present in follicular fluid during the preovulatory period, and it has been hypothesized that it plays a significant role in release of the oocyte. High concentrations of the anaphylatoxins C3a, C4a and C5a were found in human follicular fluid, the presence of these by-products indicating that complement in follicular fluid is partially activated (Perricone et al. 1990). In addition, rOGED shows the mRNA expression of several complement components, including C1qb, C3, C4, C5r1, factor h, and C1 qbp, in the ovary of the superovulated immature rat (Jo et al. 2004) (http://web5.mccs.uky.edu/kolab/rogedendo.aspx). Thus, it is clear that many immune factors, including complement, are expressed in the periovulatory ovary, strongly indicating that the expression of GPI-DAF and CD59 at this time may be serving to keep these complement components in check.
Of the three isoforms of DAF which exist in the rat, GPI-DAF was found to be the predominant form expressed in the ovary. GPI-DAF is the major form expressed in most tissues of the rat, including lung, spleen and kidney (Miwa et al. 2000). The only tissue found thus far to express high levels of TM-DAF mRNA is the testis (Miwa et al. 2000, 2001). It has been speculated that since mature sperm cannot synthesize new protein, the more stable TM protein may be needed to provide protection (Miwa et al. 2001). The ovary, like most tissues, being able to make new proteins, would not need to express TM-DAF preferentially over GPI-DAF.
We sought to discover which factors are involved in the regulation of GPI-DAF mRNA expression in the peri-ovulatory rat ovary. Since it was apparent that hCG induces GPI-DAF mRNA expression in vivo, we first investigated pathways stimulated by hCG. Perhaps the most well-known function of hCG is to activate the P4 and PG pathways, which are critical for successful ovulation (Richards et al. 1998). The anti-P4 compound RU486 and PG synthesis inhibitor indomethacin both exert anti-ovulatory effects (Richards et al. 1998), and null mutants for P4 receptor (PR) or the cyclooxygenase-2 gene are infertile due to anovulatory syndromes (Richards et al. 2002). PR is transcribed in granulosa cells of pre-ovulatory follicles in response to LH; as a nuclear receptor transcription factor, it mediates the action of P4 by controlling the transcription of specific genes (Robker et al. 2000). PG may also activate genes critical to ovulation by binding to its membrane receptor (Richards et al. 1998). The administration of RU486 and indomethacin in vivo did not block GPI-DAF mRNA expression, indicating that GPI-DAF is not one of the genes under regulation of PR or PGs. Therefore, GPI-DAF may be activated directly by hCG. Interestingly, while we have seen only a minor increase of GPI-DAF mRNA expression by PMSG (Fig. 1), GPI-DAF transcript showed an 11-fold increase in immortalized rat granulosa cells (Sasson et al. 2003). This dramatic increase may be in part due to the artificial state of the cell line, which expresses follicle-stimulating hormone receptors at 20-fold higher rates than primary cells.
To see whether or not GPI-DAF is under the direct regulation of a protein kinase cascade, we studied the effect of several factors on GPI-DAF mRNA expression in theca cells in vitro. We demonstrated that the expression of GPI-DAF in theca cells in vitro is induced by forskolin, which activates the PKA pathway (Fig. 6B). Interestingly, cotreatment of cells with either CHX+hCG or CHX+forskolin induced an even more dramatic increase in GPI-DAF than hCG or forskolin alone. This increased expression could be due to increased mRNA stability sometimes afforded by CHX or may indicate that synthesis of a new protein actually serves to inhibit induction of GPI-DAF. It was also noticeable that a significant increase in GPI-DAF expression was detected in theca cell culture without any treatment in the 6 h after isolation (Fig. 6A). Whether this is due to the stress of the cell isolation procedure or removal from its in vivo conditions, this rise may mask effects of treatment when assayed at 6 h. This result, along with the fact that inhibitors of P4 and PG did not suppress GPI-DAF induction, would indicate that hCG effects transcription of GPI-DAF via activation of the PKA cascade. This does not rule out activation of protein kinase by other receptors, such as cytokine receptors. In murine endothelial cells, DAF is induced by tumor necrosis factor-α, and this induction is PKC-dependent (Ahmad et al. 2003). However, we did not observe induction of GPI-DAF after treatment with several cytokines, nor with a PKC agonist (data not shown). The LH surge in vivo and the subsequent changes in the physical and biochemical environment surrounding the theca cells, such as the rapid increase in follicular size, angiogenesis, influx of leukocytes, and loosening of the extracelluar structure by proteinases, may contribute to the fine regulation of GPI-DAF expression in vivo.
Coupled with previous reports that complement regulators are detected in the ovary of human and other species, our findings strongly indicate that DAF may play a crucial role in preventing damage that would otherwise be induced by activation of the complement cascade, supporting the reports of a correlation between deteriorating complement function and cystic or multifollicular ovaries in women with hereditary angioedema (Perricone et al. 2000). We hypothesize that a lack of functional DAF or other complement regulators could contribute to inappropriate damage to ovarian tissue. After hundreds of ovulatory cycles, a lack of protection could lead to accumulated damage in the older female, which might cause destruction of small follicles, a decrease in follicular reserve, and possibly premature menopause. Two independent groups have generated a DAF knockout mouse (Sun et al. 1999, Lin et al. 2001); however, the phenotype of the ovary has not been reported. The DAF knockout mice are fertile (Sun et al. 1999), yet to the best of our knowledge, no long-term fertility assays have been performed. We plan to perform a long-term fertility assay to test our hypothesis that complement regulators prevent accumulated damage in aging females. We will take into account the fact that mice, unlike any other mammal, have two genes for DAF, and the DAF knockout mice are null only for Daf1. Most of the GPI-DAF product is produced from Daf1, but Daf2 is also capable of producing GPI-DAF (Harris et al. 1999).
The presence of complement in the ovary during the periovulatory period requires a delicate balance. Complement may contribute and even be necessary for successful ovulation, yet at the same time the amplifying nature of the cascade must be controlled to prevent excess inflammation and subsequent tissue damage. We believe that the presence of active inflammatory and complement factors before ovulation requires a regulatory system to protect the healthy tissue.
mRNA expression profile of GPI-DAF. (A) Expression of GPI-DAF was determined using the Rat Ovarian Gene Expression Database (rOGED), which contains DNA microarray analysis from the Affymetrix Rat Expression 230A and 230B gene chips. Immature rats were treated with 10 IU PMSG, followed by 10 IU hCG 48 h later. Tissue samples were collected at PMSG 0, 12 and 48 h and hCG 6 and 12 h. Five animals were used per time point; four ovaries (from each of four individual animals) were used for total RNA extraction, while the remaining six ovaries were used for granulosa cell isolation and residual ovarian cell collection. The experiment was repeated for whole ovary and granulosa cells. s.d. error bars are shown for whole ovary. (B) Autoradiograph for Northern blot analysis shows GPI-DAF mRNA expression increases at hCG 6 and 12 h, specifically in residual tissue. RNA from whole ovary, granulosa cells and residual tissue was collected at PMSG 0, 12 and 48 h and hCG 6 and 12 h as above. Five micrograms of RNA were separated by electrophoresis, then blots were prepared and probed with antisense GPI-DAF probe. The 28S RNA band demonstrates equal loading of lanes.
Citation: Journal of Endocrinology 186, 2; 10.1677/joe.1.06218
Characterization of isoform expression for DAF. Northern analysis revealed the existence of three transcripts for GPI-DAF. Three isoforms of DAF exist in the rat: a glycosylphosphatidylinositol (GPI), transmembrane (TM) and soluble (Sol) form. In order to determine whether the TM or soluble form are also expressed in the periovulatory ovary, Northern analysis was performed using the GPI-DAF probe, and probes specific for TM-DAF and soluble-DAF respectively. No signal was detected for TM-DAF or soluble-DAF in the ovary of PMSG 48 h, hCG 6 or 24 h primed rats by Northern blot analysis. Reverse agarose gel image of 28S RNA band shows equal amounts of RNA for each sample.
Citation: Journal of Endocrinology 186, 2; 10.1677/joe.1.06218
Expression of GPI-DAF vs TM-DAF in PMSG 48 h, and hCG 6 h ovary. (A) RT-PCR comparing relative expression levels of GPI-DAF and TM-DAF in the PMSG 48 h and hCG 6 h ovary. Primers for GPI-DAF or TM-DAF were used to amplify cDNA for 20, 25, 30 and 35 cycles. In the hCG 6 h ovary, TM-DAF PCR product is not detected until 30 cycles, while GPI-DAF is detectable after only 20 cycles. Thus TM-DAF transcript is present, albeit at very low levels. GPI-DAF and TM-DAF plasmids were used as template for control. (B) Graph depicting relative signal intensity of PCR products. In the hCG 6 h ovary, GPI-DAF PCR product shows approximately five times greater signal intensity than the TM-DAF PCR product.
Citation: Journal of Endocrinology 186, 2; 10.1677/joe.1.06218
Confirmation of GPI-DAF mRNA expression profiles by in situ hybridization. Immature 21-day-old rats were primed with PMSG (10 IU), followed by hCG (10 IU) 48 h after PMSG injection. Ovaries with intact oviducts were collected at PMSG 48 h and hCG 6, 12 and 24 h, sectioned, and hybridized with antisense probe for GPI-DAF. (A) GPI-DAF mRNA transcript is localized to the theca-interstitial cells (TIC), while no signal is visible in granulosa cells (GC). Expression peaks at hCG 6 h. (B) Whole ovary at hCG 6 h shows expression of GPI-DAF throughout the theca cell layer. (C) GPI-DAF in oviduct shows strong signal in the luminal epithelium (LE), but not in the smooth muscle layer (SM). Original magnification of slides × 10 for (A) and (C), × 2 for (B). POF, pre-ovulatory follicle.
Citation: Journal of Endocrinology 186, 2; 10.1677/joe.1.06218
GPI-DAF mRNA expression is not directly controlled by P4 or PG pathways. (A) Immature rats (21 days old) were treated with 10 IU PMSG, followed by 10 IU hCG at PMSG 48 h. One group of animals was treated with RU486 (2 mg/kg body weight) at PMSG 47 h, while another group was treated with indomethacin (1 mg/rat) at PMSG 51 h. Ovaries were harvested at both 6 and 24 h post-hCG for both RU486- and indomethacin-treated animals. Control animals were killed at PMSG 48 h, hCG 6 and 24 h. One animal was used for each time point/treatment; the experiment was performed three times. (B) Ten micrograms of total RNA were separated by electrophoresis, and blots were probed with GPI-DAF antisense probe. (C) In situ hybridization of sections from hCG 6 h, hCG 6 h+RU486, and hCG 6 h+indomethacin with GPI-DAF antisense probe. No marked decrease in mRNA transcript is seen between hCG-treated and RU486- or indomethacin-treated groups.
Citation: Journal of Endocrinology 186, 2; 10.1677/joe.1.06218
Forskolin induces GPI-DAF mRNA expression in vitro. (A) Theca cells were isolated from PMSG 48 h primed 23-day-old immature rats by the discontinuous density gradient centrifugation method and cultured up to 72 h without treatment, and the basal level of GPI-DAF expression examined (n=2). Without treatment, GPI-DAF expression increases significantly after 6 h (P=0.018). 0 vs 12 h, P=0.074; 0 vs 18 h, P=0.048; 0 vs 24 h, P=0.054; 0 vs 72 h, P=0.015. The LH-R mRNA was present in theca cells cultured for 6 h, but not in cells cultured for 18 h (not shown). For subsequent experiments, cells were treated immediately after plating, despite the rise in basal level of GPI-DAF, to have optimal LH-R responsiveness. (B) Theca cells were isolated from PMSG 48 h primed 23-day-old immature rats and treated immediately with 0.1 IU/ml hCG, 10 μg/ml CHX, 10 μM forskolin (PKA pathway activator), 20 nM PMA (PKC pathway activator) or hCG+CHX and forskolin+CHX. After 6 h, cells were harvested, RNA extracted and GPI-DAF mRNA transcript detected by semi-quantitative RT-PCR. Treatment with hCG+CHX or forskolin+CHX compared with non-treated cells was found to be significantly different from zero by one-way ANOVA analysis. L19 was used as an internal control. Relative expression is shown as the ratio of GPI-DAF to L19. The experiment was performed five times. NT, no treatment; FSK, forskolin. s.d. error bars are shown. Asterisks indicate a significant increase.
Citation: Journal of Endocrinology 186, 2; 10.1677/joe.1.06218
mRNA expression profile of CD59 in ovary. CD59, another complement regulator, was also found to be expressed in the periovulatory ovary using the rOGED database (see Fig. 1 legend). Expression is consistently high in residual tissue, and increases dramatically in granulosa cells by hCG 12 h. s.d. error bars are shown for ovary.
Citation: Journal of Endocrinology 186, 2; 10.1677/joe.1.06218
(M C Gieske and G Y Na contributed equally to this work)
The authors thank Dr Clinton Allred, Mr Chase Southard, and Mr Byungkyu Kim for help with statistical analysis.
Funding
This work was supported by grant P20 RR15592 from the National Institutes of Health and Chemyong Ko’s start-up fund from the University of Kentucky. No conflict of interest exists with regard to this work.
References
Ahmad SR, Lidington EA, Ohta R, Okada N, Robson MG, Davies KA, Leitges M, Harris CL, Haskard DO & Mason JC 2003 Decay-accelerating factor induction by tumour necrosis factor-alpha, through a phosphatidylinositol-3 kinase and protein kinase C-dependent pathway, protects murine vascular endothelial cells against complement deposition. Immunology 110 258–268.
Brannstrom M & Enskog A 2002 Leukocyte networks and ovulation. Journal of Reproductive Immunology 57 47–60.
Cole DS & Morgan BP 2003 Beyond lysis: how complement influences cell fate. Clinical Science (London) 104 455–466.
Curry TE Jr & Osteen KG 2003 The matrix metalloproteinase system: changes, regulation, and impact throughout the ovarian and uterine reproductive cycle. Endocrine Reviews 24 428–465.
Espey LL 1980 Ovulation as an inflammatory reaction – a hypothesis. Biology of Reproduction 22 73–106.
Harris CL, Rushmere NK & Morgan BP 1999 Molecular and functional analysis of mouse decay accelerating factor (CD55). Biochemical Journal 341 821–829.
Hinchliffe SJ, Spiller OB, Rushmere NK & Morgan BP 1998 Molecular cloning and functional characterization of the rat analogue of human decay-accelerating factor (CD55). Journal of Immunology 161 5695–5703.
Hourcade D, Liszewski MK, Krych-Goldberg M & Atkinson JP 2000 Functional domains, structural variations and pathogen interactions of MCP, DAF and CR1. Immunopharmacology 49 103–116.
Jo M, Gieske MC, Payne CE, Wheeler-Price SE, Gieske JB, Ignatius IV, Curry TE Jr & Ko C 2004 Development and application of a rat ovarian gene expression database. Endocrinology 145 5384–5396.
Ko C & Park-Sarge OK 2000 Progesterone receptor activation mediates LH-induced type-I pituitary adenylate cyclase activating polypeptide receptor (PAC(1)) gene expression in rat granulosa cells. Biochemical and Biophysical Research Communications 277 270–279.
Ko C, In YH & Park-Sarge OK 1999 Role of progesterone receptor activation in pituitary adenylate cyclase activating polypeptide gene expression in rat ovary. Endocrinology 140 5185–5194.
Ko C, Grieshaber NA, Ji I & Ji TH 2003 Follicle-stimulating hormone suppresses cytosolic 3,5,3′-triiodothyronine-binding protein messenger ribonucleic acid expression in rat granulosa cells. Endocrinology 144 2360–2367.
Kuby J 1997 Immunology. New York: WH Freeman and Company.
Kumar S, Vinci JM, Pytel BA & Baglioni C 1993 Expression of messenger RNAs for complement inhibitors in human tissues and tumors. Cancer Research 53 348–353.
Lin F, Fukuoka Y, Spicer A, Ohta R, Okada N, Harris CL, Emancipator SN & Medof ME 2001 Tissue distribution of products of the mouse decay-accelerating factor (DAF) genes. Exploitation of a Daf1 knock-out mouse and site-specific monoclonal antibodies. Immunology 104 215–225.
Liszewski MK, Farries TC, Lublin DM, Rooney IA & Atkinson JP 1996 Control of the complement system. Advances in Immunology 61 201–283.
Magoffin DA & Erickson GF 1988 Purification of ovarian theca-interstitial cells by density gradient centrifugation. Endocrinology 122 2345–2347.
Mead R, Hinchliffe SJ & Morgan BP 1999 Molecular cloning, expression and characterization of the rat analogue of human membrane cofactor protein (MCP/CD46). Immunology 98 137–143.
Miwa T & Song WC 2001 Membrane complement regulatory proteins: insight from animal studies and relevance to human diseases. International Immunopharmacology 1 445–459.
Miwa T, Okada N & Okada H 2000 Alternative exon usage in the 3′ region of a single gene generates glycosylphosphatidylinositol-anchored and transmembrane forms of rat decay-accelerating factor. Immunogenetics 51 129–137.
Miwa T, Sun X, Ohta R, Okada N, Harris CL, Morgan BP & Song WC 2001 Characterization of glycosylphosphatidylinositol-anchored decay accelerating factor (GPI-DAF) and transmembrane DAF gene expression in wild-type and GPI-DAF gene knockout mice using polyclonal and monoclonal antibodies with dual or single specificity. Immunology 104 207–214.
Nauta AJ, Roos A & Daha MR 2004 A regulatory role for complement in innate immunity and autoimmunity. International Archives of Allergy and Immunology 134 310–323.
Ness RB, Grisso JA, Cottreau C, Klapper J, Vergona R, Wheeler JE, Morgan M & Schlesselman JJ 2000 Factors related to inflammation of the ovarian epithelium and risk of ovarian cancer. Epidemiology 11 111–117.
Norman RJ & Brannstrom M 1996 Cytokines in the ovary: pathophysiology and potential for pharmacological intervention. Pharmacology and Therapeutics 69 219–236.
Oglesby TJ, Longwith JE & Huettner PC 1996 Human complement regulator expression by the normal female reproductive tract. Anatomical Record 246 78–86.
Perricone R, de Carolis C, Moretti C, Santuari E, de Sanctis G & Fontana L 1990 Complement, complement activation and anaphylatoxins in human ovarian follicular fluid. Clinical and Experimental Immunology 82 359–362.
Perricone R, Pasetto N, De Carolis C, Vaquero E, Noccioli G, Panerai AE & Fontana L 1992 Cystic ovaries in women affected with hereditary angioedema. Clinical and Experimental Immunology 90 401–404.
Perricone R, De Carolis C, Giacomello F, Giacomelli R, De Sanctis G & Fontana L 2000 Impaired human ovarian follicular fluid complement function in hereditary angioedema. Scandinavian Journal of Immunology 51 104–108.
Richards JS, Russell DL, Robker RL, Dajee M & Alliston TN 1998 Molecular mechanisms of ovulation and luteinization. Molecular and Cellular Endocrinology 145 47–54.
Richards JS, Russell DL, Ochsner S & Espey LL 2002 Ovulation: new dimensions and new regulators of the inflammatory-like response. Annual Review of Physiology 64 69–92.
Robker RL, Russell DL, Espey LL, Lydon JP, O’Malley BW & Richards JS 2000 Progesterone-regulated genes in the ovulation process: ADAMTS-1 and cathepsin L proteases. PNAS 97 4689–4694.
Runesson E, Ivarsson K, Janson PO & Brannstrom M 2000 Gonadotropin- and cytokine-regulated expression of the chemokine interleukin 8 in the human preovulatory follicle of the menstrual cycle. Journal of Clinical Endocrinology and Metabolism 85 4387–4395.
Sasson R, Dantes A, Tajima K & Amsterdam A 2003 Novel genes modulated by FSH in normal and immortalized FSH-responsive cells: new insights into the mechanism of FSH action. FASEB Journal 17 1256–1266.
Spiller OB, Hanna SM & Morgan BP 1999 Tissue distribution of the rat analogue of decay-accelerating factor. Immunology 97 374–384.
Sun X, Funk CD, Deng C, Sahu A, Lambris JD & Song WC 1999 Role of decay-accelerating factor in regulating complement activation on the erythrocyte surface as revealed by gene targeting. PNAS 96 628–633.
Ushigoe K, Irahara M, Fukumochi M, Kamada M & Aono T 2000 Production and regulation of cytokine-induced neutrophil chemoattractant in rat ovulation. Biology of Reproduction 63 121–126.
Walport MJ 2001 Complement. First of two parts. New England Journal of Medicine 344 1058–1066.
Wong KH, Negishi H & Adashi EY 2002 Expression, hormonal regulation, and cyclic variation of chemokines in the rat ovary: key determinants of the intraovarian residence of representatives of the white blood cell series. Endocrinology 143 784–791.
