- Email: [email protected]
Suppression of lymphocyte proliferation by ovarian cavity fluid from the viviparous fish Neoditrema ransonnetii (Perciformes; Embiotocidae)
Fish & Shellfish Immunology 27 (2009) 549–555
Contents lists available at ScienceDirect
Fish & Shellfish Immunology journal homepage: www.elsevier.com/locate/fsi
Suppression of lymphocyte proliferation by ovarian cavity fluid from the viviparous fish Neoditrema ransonnetii (Perciformes; Embiotocidae) Erina Saito*, Osamu Nakamura, Hiroyuki Yamada, Shigeyuki Tsutsui, Tasuku Watanabe School of Marine Biosciences, Kitasato University, Ofunato, Iwate 022-0101, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 10 December 2008 Received in revised form 13 July 2009 Accepted 14 July 2009 Available online 21 July 2009
As the fetus expresses paternal major histocompatibility complex molecules, viviparous vertebrates require sophisticated mechanisms to modulate maternal immunology to ensure successful pregnancy. We anticipated that ovarian cavity fluid (OCF) is likely to feature significantly in the modulation of ovarian cavity immunology. Consequently, we examined the effects of OCF upon leukocyte function in Neoditrema ransonnetii. OCF did not affect phagocytosis or superoxide production by phagocytes. However, OCF suppressed lymphocyte proliferation induced by ConA almost completely. As OCF contained PGE2 at high levels during late pregnancy, we also investigated the effect of PGE2 upon lymphocyte expansion. PGE2 exhibited negative effects upon lymphocyte mitogenesis in a dosedependent manner (10–1000 ng/ml). PGE2 significantly suppressed lymphocyte proliferation when present at levels equivalent to that seen in OCF (30.2 16.1 w 185.4 107.4 ng/ml). Data indicate that PGE2 is one of the key modulatory molecules of the maternal immune system ensuring successful pregnancy in this viviparous species. Ó 2009 Elsevier Ltd. All rights reserved.
Keywords: Viviparity Ovarian cavity fluid Immunomodulation Lymphocyte proliferation Prostaglandin E2 Pregnancy Surfperch
1. Introduction Viviparity requires highly sophisticated maternal control mechanisms in order to modulate immune responses effectively because the spermatozoa and fetuses express paternal major histocompatibility complex (MHC) antigens. It is essential for the success of internal fertilization and maintenance of pregnancy that local immune reactions are regulated. A number of studies have attempted to elucidate the immuno-modulating mechanisms which underlie pregnancy for humans and other mammalian species, and these studies have been largely successful. Such mechanisms include the restricted expression of MHC molecules on the trophoblast [1–4], presence of multiple immunoregulatory factors (e.g., cytokines, hormones, and prostaglandin E2 (PGE2)) [5–8], and the distinct distribution of immune cells across the maternal–fetal interface [9–15]. In stark contrast with mammalian species, this puzzle has been left almost unexplored in non-mammalian viviparous species, despite the fact that viviparity is widespread among vertebrates, with the exception of Aves and Agnatha. There are over one thousand species of viviparous fish [16]. Fish are the first vertebrate group to have developed MHC and T cell receptors (TCR), and, as * Corresponding author. Tel.: þ81 192 44 1907; fax: þ81 192 44 2125. E-mail address: [email protected] (E. Saito). 1050-4648/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.fsi.2009.07.006
a result, exhibit strict immune-rejection against non-self transplants [17–22]. Viviparous fishes are, therefore, the oldest creatures to have faced conflict between the immune system, internal fertilization and fetal growth. Understanding how fish have resolved this conflict would reveal vital information regarding the evolutionary process of viviparity in vertebrates. Embiotocidae fish, also known as ‘surfperch’, possess some unique characteristics among viviparous teleosts; their gestational period is considerably long (more than 6 months), the fetus has very limited yolk reserves resulting in almost complete dependence on maternally supplied nutrients. The fish do not develop a placenta. Instead, fetuses continuously receive maternal secretions, known as the ovarian cavity fluid (OCF). Neoditrema ransonnetii is a common Embiotocidae fish found in Japan. Copulating behavior is observed from September to December, in Iwate, Northern Japan. Internal fertilization appears to occur from late December through to early January. Fetuses retained in the ovarian cavity ingest and vigorously absorb OCF via their hypertrophied hindgut [23]. From late July to early August, newborn are delivered. Through microscopic observation, we previously found that a large number of leukocytes exist in the ovarian cavity [24]. This indicates that inseminated sperm cells and developing fetuses are not immunologically sequestered from the maternal immune system in surfperch. Vigorous phagocytosis of the inseminated sperm cells by macrophages is evident in the ovarian lumen after coitus, and even
550
E. Saito et al. / Fish & Shellfish Immunology 27 (2009) 549–555
during the gestation period [24]. Nevertheless, neither an apparent immune response nor a semi-allogenic rejection appears to occur. Throughout the reproductive cycle, more than 90% of observed leukocytes in the ovarian lumen are monocytes/macrophages, while lymphocytes invariably represent only a small population. These facts suggest the presence of local mechanisms suppressing lymphocyte influx or proliferation. We therefore expected the OCF to play key roles in the immune modulation of the ovarian cavity in order to maintain pregnancy in surfperch. In the current study, we examined the immuno-modulating effect of OCF upon phagocytic functions and lymphocyte proliferation. In addition, we also studied the effect of PGE2, a well-known inflammatory mediator and immuno-modulating agent, because we found that OCF contained high levels of PGE2. 2. Materials and methods 2.1. Fish Fish were collected by angling at Okirai Bay, Iwate, Japan, and were maintained under natural conditions with running sea water in fiber-reinforced plastic tanks at Kitasato University. Fish were fed a commercial diet for marine fish (Fuji-Seifun Co., Japan). To avoid any possible influences that can be induced by pregnancy-related factors, the cells used in all experiments were obtained from the males. 2.2. Collection of plasma and OCF Fishes were anesthetized with 2-phenoxyethanol, and blood collected using heparinized syringes with a 26G needle. Following centrifugation at 700 g for 15 min at 4 C, the supernatants were harvested and stored at 80 C to await analysis. Following laparotomy, the ovaries were dissected and the OCF collected by a micropipette. Following centrifugation, the supernatants were preserved at 80 C and sterilized with a syringe filter unit (pore size, 0.20 mm; ADVANTEC, Japan) before use. 2.3. Cell preparation Blood was drawn aseptically from the caudal vein of anesthetized fish into a syringe containing 1000 IU/ml (the final dilution: 125 IU/ml) of heparin and immediately mixed with 5 volumes of 1 Dulbecco’s phosphate-buffered saline (D-PBS) (Gibco, USA). The diluted blood was then layered on 45% PercollÔ (GE Healthcare, USA) prior to centrifugation at 400 g for 40 min at 20 C. The peripheral blood leukocytes (PBL) layer at the interface was harvested and washed 3 times before use in the following experiments. Following laparotomy, the head kidney was removed and pushed through a stainless mesh in D-PBS. The resulting cell suspension was placed on 35/51% PercollÔ and centrifuged. The macrophage-enriched cells from the 35/51% PercollÔ interface were recovered, washed with 0.01 M PBS, and resuspended in the RPMI 1640 medium (Nissui Pharmaceutical, Japan). These cells are referred to as head kidney leukocytes (HKLs) hereafter. 2.4. Phagocytosis assay HKLs (2 106 cells) in RPMI 1640 were allowed to adhere to a culture coverglass (Matsunami Glass, Japan) for 24 h at 15 C, and non-adherent cells were removed by washing with 0.01 M PBS. Cells were then maintained with 150 ml of OCF or RPMI 1640 at 15 C for 6 h. Following washing, latex beads (L.B, 0.75 mm, Polysciences Inc, USA), opsonized with N. ransonnetii serum, were
added to the well along with the RPMI 1640 medium. After 1 h of incubation at 15 C and washing, the cells on the cover slip were fixed with methanol and stained with 40 -6-diamino-2-phenylindole. The number of phagocytic cells that ingested at least one particle was counted microscopically and the phagocytic activity was expressed as a percentage of the total cell counts. Opsonic activity of OCF was also examined. L.B. were treated with OCF, serum, or RPMI 1640 medium for 1 h at room temperature. Following washing, these were resuspended in RPMI 1640 medium, and added to HKL monolayers on the cover slip. Phagocytic activity was measured as described above.
2.5. Intracellular production of superoxide anions in the macrophages Superoxide anion production by HKLs was determined according to the method described by Chung and Secombes [25]. In brief, the HKLs (1 106 cells/well) were plated onto 96-well microtiter plates (AGC Techno glass, Japan) and allowed to adhere for 24 h at 15 C. After washing with RPMI 1640, the macrophage monolayers were treated with 80 ml of OCF, male serum, or RPMI 1640 for 24 h. After another round of washing, the cells were incubated with 100 ml of NBT solution (1 mg/ml) in RPMI 1640 with or without Zymosan A (2.5 107 particles/well) (Sigma, USA) opsonized with N. ransonnetii serum. After 1 h, the NBT solution was removed and the cells were fixed with methanol. The reduced formazan within the macrophages was solubilized in KOH/DMSO for 5 min and the optical density was measured by a microplate reader (Model 550; BioRad, USA) at 620 nm. 2.6. Cell proliferation assay Isolated PBLs were resuspended in OCF or Leibovitz’s L-15 medium (Gibco) containing 100 IU/ml penicillin and 100 mg/ml streptomycin. The cell suspensions were incubated in 96-well microplates with or without 1 mg/well of concanavalin A (ConA, Sigma) for 72 h at 15 C. Twenty-four hours before the end of cultivation, 5-bromo-20 -deoxyuridine (BrdU, final concentration: 10 mM) was added to the culture. BrdU uptake into nuclear DNA was measured using the Cell Proliferation ELISA kit (GE Healthcare) according to the manufacturer’s instructions.
2.7. Characterization of the proliferating cells To characterize the cells proliferating in response to ConA, we centrifuged and smeared a few of these cells onto glass slides. Following acetone fixation and blocking with 1% FBS, the cells were incubated with anti-IgM rabbit antiserum ( 10,000) [26] for 2 h at room temperature. After washing, the cells were reacted with antirabbit IgG goat IgG conjugated with Alexa Fluor 633 (Funakoshi, Japan) for 1 h. After washing, DNA was fragmented using DNase (Promega, USA) for 1 h at 37 C. The cells were washed again and then exposed to 10 mg/ml of FITC-labeled anti-BrdU mouse mAb (Biomeda Corporation, USA) for 2 h. Lectin staining with ConA was also performed in order to identify the ConA-binding cells. PBLs were fixed with 4% paraformaldehyde and smeared by cytospin. After blocking of endogenous biotin with the avidin/biotin blocking kit (Vector Laboratories, USA), cells were treated with 5 mg/ml of biotin-labeled ConA (Sigma) for 1 h and then incubated with 100 mg/ml of FITC-conjugated streptavidine (Sigma) for 30 min. All fluorescent signals were captured by a confocal laser-scanning microscope (LSM-510META; Carl Zeiss, Germany).
E. Saito et al. / Fish & Shellfish Immunology 27 (2009) 549–555
2.8. Hemagglutination assay To confirm whether OCF interrupted the activity of the mitogen, we examined the hemagglutination activity of ConA in the presence of OCF. First, the lectin activity of OCF was assessed. OCF itself clumped rabbit red blood cells (RRBCs), and this activity was inhibited by 0.2 M L-()-rhamnose or L-()-fucose. Therefore, 0.2 M rhamnose was added to OCF or the L-15 medium and pre-incubated for 1 h at room temperature. ConA (40 mg/ml) was serially diluted with these solutions on 96-well V-bottom microtiter plates (Kartell, Italy) and mixed with the RRBC suspension. 2.9. Effect of PGE2 on PBL proliferation induced by ConA PGE2 concentrations in the plasma and OCF were measured using a commercial ELISA kit (Cayman Chemicals, USA) according to the manufacturer’s instructions. The antibody in this kit shows crossreactivity with PGE3 (43%) and PGE1 (20%). To confirm whether the substances measured in the ELISA was PGE2 or other E series PGs, a low molecular mass fraction of OCF was prepared by ultrafiltration using Vivaspin 5 k (Sartorius Stedim Biotech, Germany) and analyzed by MALDI TOF-MS (Shimadzu biotech, Japan). PBLs were suspended in L-15 medium and plated onto 96-well plates as described above. PGE2 solutions (0.01–1000 ng/ml) containing 0.01% ethanol and 20 mg/ml of ConA were added to each well (50 ml/well). Control cultures were tested with regard to the effects of ethanol at the same final dilution (0.01%). After 72 h of incubation, cell proliferation was determined by the BrdU assay as described earlier. 2.10. Statistical analysis All numerical data are presented as mean SD. Data from the phagocytosis assay and quantification of PGE2 were analyzed by paired and unpaired t tests, respectively. Other data were analyzed using repeated measures ANOVA followed by Bonferroni pairwise comparisons or Dunnett’s multiple comparison test. Differences were considered as statistically significant at the 95% confidence level (p < 0.05). 3. Result 3.1. Effect of OCF on phagocytic functions The phagocytic activity of renal macrophages against microspheres that had been opsonized by serum was not affected by the presence of OCF (Fig. 1). The phagocytic rate of HKLs in the OCF did
Fig. 1. Effect of OCF upon phagocytosis. After HKLs were treated with OCF or RPMI 1640, phagocytic activity was assessed. Data are expressed as mean SD (n ¼ 3) of the percentage of cells phagocytizing L.B.
551
not differ from that in the medium. On the other hand, OCF exhibited significant opsonic activity for HKLs that was nearly equal to that of the serum (Fig. 2). On stimulation with opsonized zymosan, HKLs in the OCF produced superoxide anions at levels almost equal to that in the male serum, but higher than that observed in the medium (Fig. 3). 3.2. Effect of OCF on lymphocyte proliferation In L-15 medium, PBLs underwent proliferation upon stimulation with ConA within 72 h of incubation. This induced proliferation was almost completely abolished in the OCF (Fig. 4). Although ConA is known to act as a T cell mitogen in mammals, the type of cells that respond to ConA in fish has yet to be clarified. Thus, we morphologically characterized cells incorporating BrdU or bearing binding sites for ConA. As shown in Fig. 5A and B , both these cells displayed the typical features of small-sized lymphocytes with a large nucleo-cytoplasmic ratio. When the cells were double stained with anti-BrdU and anti-IgM antibodies, nearly half of the cells incorporating BrdU were IgM negative (Fig. 5A), but the other half was IgM positive (Fig. 5B). Negative controlled cells that were treated without anti-BrdU mAb and anti-IgM antiserum or biotin-labeled ConA were not stained (data not shown). These results indicate that the cells undergoing proliferation upon ConA stimulation were lymphocytes-like cells. To ascertain that the suppression of mitogenesis was a direct result of the inhibitory effects of OCF upon lymphocytes but not from its interference with the mitogen, we further investigated the agglutination activity of ConA against RRBCs in the presence of OCF. Having performed three times of similar experiments, the results showed that the agglutination titer of ConA in the OCF was not significantly different from that in the medium (Table 1). 3.3. PGE2 is the proliferation inhibitory factor in the ovarian cavity PGE2 concentration was measured in plasma and OCF from females. The levels in the plasma of pregnant females in July, i.e., a few weeks before parturition, tended to increase but not significantly. In addition, there was little difference between pregnant and non-pregnant female in plasma PGE2 levels. On the other hand, levels in the OCF in April were 185.4 107.4 ng/ml. This was significantly higher than those of other periods, although the values in April markedly varied between individuals. In July, the PGE2 concentration in the OCF was 30.2 16.1 ng/ml, approximately 10fold higher than in the plasma (Fig. 6). Because the antibody in this
Fig. 2. Opsonic activity of OCF. Phagocytes were brought into contact with L.B. pretreated with OCF, serum, or the RPMI 1640 medium, and incubated for 1 h. Data are expressed as mean SD (n ¼ 5) of the percentage of cells phagocytizing L.B. *, significant at p < 0.05; **, significant at p < 0.01.
552
E. Saito et al. / Fish & Shellfish Immunology 27 (2009) 549–555
Fig. 3. Effect of OCF on superoxide anion production. After the macrophages were incubated in OCF, male serum, or RPMI 1640 medium, intracellular superoxide anions were measured by NBT reduction. Data are expressed as mean SD (n ¼ 5). *, significant at p < 0.05.
kit has cross-reactivity with other E series PGs such as PGE1 and PGE3, OCF was analyzed by MALDI TOF-MS to confirm whether the substances detected in this study were PGE2 or other E series PGs. A peak was detected at almost the same molecular mass as PGE2 (352.5) (Fig. 7). However, peaks corresponding to the molecular masses of PGE1 (354.5) and PGE3 (350.5) were not detected. Therefore, it was concluded that the molecules measured by the kit in this study were PGE2, at least the case of the OCF in July. It was not possible to analyze OCF before July because its amount was too small. We then examined the effect of PGE2 upon lymphocyte proliferation. As shown in Fig. 8, PGE2 strongly inhibited the ConAinduced proliferation of PBLs. This effect was dose-dependent and significant suppression was observed at concentrations from 10 to 1000 ng/ml. 4. Discussion In the current study, we demonstrate that the ovarian cavity fluid (a histotrophe secreted by ovarian tissues) from surfperch exerted a potent inhibitory effect upon lymphocyte-like cells proliferation induced by ConA (Fig. 4) but did not suppress phagocytic functions. Furthermore, we demonstrated that the suppressive activity of the OCF upon those cells was ascribed to, at
Fig. 5. Immunofluorescent detection of proliferating cells and ConA-binding cells. Cells incorporating BrdU were stained with anti-IgM antibody and FITC-labeled anti-BrdU Ab (A). A1 is BrdU (green) single positive cells. A2 is IgM (purple) and BrdU double positive cells. ConA-binding cells were detected by biotin-labeled ConA and FITCconjugated streptavidine (B). Scale bar ¼ 5 mm.
least in part, PGE2 that had accumulated in the OCF in high levels (Fig. 8). ConA acts as a mitogen for mammalian T cells and has shown mitogenic activity for lymphocytes in several species of fish [27–32]. Though lymphocyte cell markers have been elucidated in some fish species, such as carp and sea bass [33–36], it is still difficult to distinguish T cells, B cells, and NK cells in the majority of fish species, including N. ransonnetii, because of unidentified those markers. Accordingly, it is unclear which types of lymphocytes are capable of responding to ConA stimulation in fish. The present study tried to identify the properties of the proliferating cells using double staining with anti-IgM antiserum and an anti-BrdU antibody. It was found that two types of cells, which were IgMþ and IgM, existed in the cells incorporating BrdU (Fig. 5). However, it was doubtful whether this anti-IgM does not react with T cells because about 50% of thymocytes can be stained using this antibody (data not shown). Therefore, the specific cell types could not be identified in the current study. However, they showed typical lymphocyte-like morphology, and at least half of them were IgM negative. Irrespective of which type of cells these are, it is important to elucidate which specific types of cells (Th, Tc, or others) can be suppressed by PGE2. Consequently, characterization of these cells will be the main focus in our future research efforts.
Table 1 The effected of OCF on agglutination of RRBC by ConA. ConA concentration (mg/ml) Fig. 4. Effect of OCF on PBL proliferation. The isolated PBLs were resuspended in OCF or L-15, with or without 1 mg/well of ConA. After 72 h, cell proliferation was measured on the basis of BrdU uptake into nuclear DNA. Data are expressed as mean SD (n ¼ 5). *, significant at p < 0.05.
OCF L-15
20
10
5
2.5
1.25
0
þ þ
þ þ
þ, agglutination; , no agglutination (n ¼ 3).
E. Saito et al. / Fish & Shellfish Immunology 27 (2009) 549–555
553
Fig. 8. Effect of PGE2 on PBL proliferation. PGE2 added to PBL at a dose of 0.01– 1000 ng/ml with 20 mg/ml ConA. After 72 h of incubation, cell proliferation was determined. Data are expressed as mean SD (n ¼ 5). *, significant at p < 0.05; **, significant at p < 0.01.
Fig. 6. PGE2 concentrations in the plasma and OCF during late pregnancy. A is the levels in plasma and B is in OCF. PGE2 concentrations were measured by ELISA. Data are expressed as mean SD (n ¼ 5). *, significant at p < 0.05.
It may be possible to explain the results shown by Fig. 4 in an alternative manner; OCF may inhibit mitogenesis not by suppressing the responsiveness of lymphocyte-like cells but rather, by interfering with the activity of the mitogen. To examine this possibility, we tested the agglutination activity of ConA against RRBCs in the OCF. As shown in Table 1, the agglutination titer of ConA in the OCF was not reduced compared with that of the medium. Therefore, we concluded that OCF could depress the proliferating ability of ConA-responding cells without interrupting the mitogen. This property of OCF may partially explain why the
Fig. 7. Characterization of molecules included in OCF by MALDI TOF-MS analysis. Low molecular fraction of OCF in July were analyzed by MALDI TOF-MS.
number of lymphocytes is not increased under the existence of immunologically foreign sperm cells and the embryos in the ovarian cavity. PGE2, an arachidonic acid derivative, is a potent bioactive substance involved in a diverse range of physiological and pathological phenomena, including salt and water excretion, bone resorption, gastric acid secretion, fever, mediation of pain, and the stimulation of inflammatory responses [37–41]. In mammals, PGE2 has been known to possess immunosuppressive activity as well; it inhibits T cell proliferation, lymphokine production, and cytotoxicity [42–48]. Although the inflammatory and immunosuppressive functions of PGE2 have not been well elucidated in fish, Secombes et al. reported that PGE2 inhibited mitogenesis in rainbow trout leukocytes in response to PHA-P [49]. In this study, it was found that OCF contained PGE2 in much higher concentrations than in plasma (Fig. 6). To ascertain whether other E series PGs were also present, OCF was analyzed by MALDI TOF-MS. As a result, a peak was detected at the appropriate position for PGE2, but neither PGE1 nor PGE3 were detected (Fig. 7). These results indicated that PGE2 exerts a significant impact upon the maintenance of gestation in N. ransonnetii. Consequently, we demonstrated that PGE2 inhibited lymphocyte proliferation in a dose-dependent manner; PGE2 significantly suppressed lymphocytic mitogenesis at concentrations from 10 to 1000 ng/ml. The mean concentration of PGE2 in the OCF in gravid females was much higher than that in the plasma in pregnant and non-pregnant females. It was not possible to determine PGE2 levels in the OCF of non-pregnant fish, because the volume of the fluid was too small to collect. These results indicated that PGE2 plays a role as an immuno-suppressor in the ovarian cavity during pregnancy. In the current study, we used OCF only from April to July. This was because it was difficult to collect sufficient amounts of OCF from the mother in the early stages of conception. The gestation period of N. ransonnetii is rather long, and the nature of OCF appears to change significantly during this period. OCF appears more dense and proteinaceous during April, the mid stage of gestation [26]. Therefore, PGE2 levels in the OCF may vary with the progression of pregnancy. In mammals, PGE2 is involved in multiple reproductive events such as ovulation, fertilization, implantation, maintenance of pregnancy, and parturition [50–54]. Thus, other functions of
554
E. Saito et al. / Fish & Shellfish Immunology 27 (2009) 549–555
PGE2 in surfperch reproduction should also be considered, and decreased levels of PGE2 in the OCF during later stages may be a transient phenomenon, possibly associated with parturition. Further research should investigate the level of PGE2 in OCF during other stages of gestation to understand the roles of this multifunctional mediator. On the other hand, one should assume that other immunomodulating substances, like sex hormones and suppressive cytokines, are also involved in the complex mechanisms leading to successful pregnancy in viviparous fish. Each of these additional molecules may play more significant roles in the immune regulation of earlier stages, if the OCF present during those periods contains reduced amounts of PGE2, as discussed above. In all experiments in the present study, we used leukocytes obtained from the head kidney and peripheral blood of male fish. Female N. ransonnetii exhibits a long mating and gestation period, and the period when females participate in neither copulation nor gestation lasts only around one month. Leukocyte responsiveness may also vary concomitant with female reproductive cycle. Therefore we needed to use males, in order to exclude any possible influences of reproduction-related factors which might have already affected maternal leukocytes. In contrast to its suppressive effect on lymphocytes, OCF did not interfere with phagocytosis or superoxide production by phagocytes. In addition, OCF possessed opsonic activity almost equal to that of the serum. Although opsonic molecules in the OCF have yet to be identified, our former study showed that OCF contained IgM, though levels were much lower than that in the plasma [26]. In addition, lectin, which was suggested to be present in OCF in the current study, may also act as an opsonin. These results indicate that the leukocyte functions associated with phagocytosis are maintained normally in the ovarian cavity of N. ransonnetii even during the gestational period. Because ovarian cavity is separated from the external environment only by a short oviduct, fetuses and ovarian epithelia may be easily threatened by invasion of microorganisms. Therefore, it seems reasonable that phagocytic functions are retained normally for the protection of fetal and maternal tissues, as in the case of mammalian genital tracts in gravid females [13,55,56]. In summary, we have provided evidence herein for the existence of immuno-modulating mechanisms during pregnancy and in particular, one specific candidate molecule, PGE2, in a viviparous fish, N. ransonnetii. Since severe conflicts between the adaptive immune system and viviparity have arisen along with the emergence of viviparous fish, the fact that surfperch exploit PGE2 as an immunomodulator for successful pregnancy would provide a further basis for understanding the evolutionary process of viviparity in vertebrates. Acknowledgements The authors thank Prof. Shunsuke Moriyama (Kitasato University) for his expertise with MALDI TOF-MS. References [1] Sargent IL. Maternal and fetal immune responses during pregnancy. Exp Clin Immunogenet 1993;10:85–102. [2] Murphy SP, Choi JC, Holtz R. Regulation of major histocompatibility complex class II gene expression in trophoblast cells. Reprod Biol Endocrinol 2004;2:52. [3] Bacon SJ, Ellis SA, Antczak DF. Control of expression of major histocompatibility complex genes in horse trophoblast. Biol Reprod 2002;66:1612–20. [4] Carter J. The maternal immunological response during pregnancy. Oxf Rev Reprod Biol 1984;6:47–128. [5] Parhar RS, Kennedy TG, Lala PK. Suppression of lymphocyte alloreactivity by early gestational human decidua. I. Characterization of suppressor cells and suppressor molecules. Cell Immunol 1988;116:392–410.
[6] Lala K, Kennedy TG, Parhar RS. Suppression of lymphocyte alloreactivity by early gestational human decidua. II. Characterization of suppressor mechanisms. Cell Immunol 1988;116:411–22. [7] Piccinni MP. T-cell cytokines in pregnancy. AJRI 2002;47:289–94. [8] Jones RL, Hannan NJ, Kaitu’u TJ, Zhang J, Salamonsen LA. Identification of chemokines important for leukocyte recruitment to the human endometrium at the times of embryo implantation and menstruation. J Clin Endocrinol Metab 2004;89:6155–67. [9] Lawrence R, Church JA, Richards W, Borzy M. Immunological mechanisms in the maintenance of pregnancy. Ann Allergy 1980;44:166–73. [10] Vince GS, Johnson PM. Leucocyte populations and cytokine regulation in human uteroplacental tissues. Biochem Soc 2000;28:191–5. [11] Ku¨hnert M, Strohmeier R, Stegmu¨ller M, Halberstadt E. Changes in lymphocyte subsets during normal pregnancy. Eur J Obstet Gynecol Reprod Biol 1998;76:147–51. [12] Trundley A, Moffett A. Human uterine leukocytes and pregnancy. Tissue Antigens 2004;63:1–12. [13] Hunt JS. Immunologically relevant cells in the uterus. Biol Reprod 1994;50:461–6. [14] Milne SA, Henderson TA, Kelly RW, Saunders PT, Baird DT, Critchley HO. Leukocyte populations and steroid receptor expression in human firsttrimester decidua; regulation by antiprogestin and prostaglandin E analog. J Clin Endocrinol Metab 2005;90:4315–21. [15] Ozenci CC, Korgun ET, Demir R. Immunohistochemical detection of CD45þ, CD56þ, and CD14þ cells in human decidua during early pregnancy. Early Pregnancy 2001;5:164–75. [16] Wourms JP. Viviparity: the maternal-fetal relationship in fishes. Am Zool 1981;21:473–515. [17] McKinney EC, McLeod TF, Sigel MM. Allograft rejection in a holostean fish, Lapososteus Platyrhincus. Dev Comp Immunol 1981;5:65–74. [18] Kaastrup P, Stet RJ, Tigchelaar AJ, Egberts E, van Muiswikel WB. A major histocompatibility locus in fish: serological identification and segregation of transplantation antigens in the common carp (Cyprinus carpio, L.). Immunogenetics 1989;30:284–90. [19] Nagelar JJ, Cloud JG, Wheeler PA, Thorgaard GH. Testis transplantation in male rainbow trout (Onchorynchus mykiss). Biol Reprod 2001;64:644–6. [20] Cardwell TN, Sheffer RJ, Hedrick PW. MHC variation and tissue transplantation in fish. J Hered 2001 Jul-Aug;92:305–8. [21] Wright Jr JR, Pohajdak B, Xu BY, Leventhal JR. Piscine islet xenotransplantation. ILAR J 2004;45:314–23. [22] Romano N, Baldassini MR, Buonocore F, Picchietti S, Mastrolia L, Abelli L. In vivo allograft rejection in a bony fish Dicentrarchus labrax (L): characterization of effector lymphocytes. Cell Tissue Res 2005 Sep;321:353–63. [23] Nakamura O, Tazumi Y, Muro T, Yasuhara Y, Watanabe T. Active uptake and transport of protein by the intestinal epithelial cells in embryo of viviparous fish, Neoditrema ransonnetii (Perciformes: Embiotocidae). J Exp Zool 2004;301A:38–48. [24] Tazumi Y, Nakamura O, Watanabe T. Intraovarioan cavity leucocytes of viviparous fish, Neoditrema ransonnetii (Perciformes, Embiotocidae). Zool Sci 2004;21:739–46. [25] Chung S, Secombes CJ. Analysis of events occurring within teleost macrophages during the respiratory burst. Comp Biochem Physiol 1988;89B:539–44. [26] Nakamura O, Kudo R, Aoki H, Watanabe T. IgM secretion and absorption in the materno-fetal interface of a viviparous teleost, Neoditrema ransonnetii (Perciformes; Embiotocidae). Dev Comp Immunol 2006;30:493–502. [27] Caspi RR, Shahrabani R, Kehati-Dan T, Avtalion RR. Heterogeneity of mitogenresponsive lymphocytes in carp (Cyprinus carpio). Dev Comp Immunol 1984;8:61–70. [28] Le Morvan-Rocher C, Troutaud D, Deschaux P. Effects of temperature on carp leukocyte mitogen-induced proliferation and nonspecific cytotoxic activity. Dev Comp Immunol 1995;19:87–95. [29] Van der Heijden MH, Booms GH, Tanck MW, Rombout JH, Boon JH. Influence of flumequine on in vivo mitogen responses on European eel (Anguilla anguilla L., 1758) lymphoid cells. Vet Immunol Immunopathol 1995;47:143–52. [30] Siwicki AK, Pozet F, Morand M, Volatier C, Terech-Majewska E. Effects of iridovirus-like agent on the cell-mediated immunity in sheatfish (Silurus glanis)dan in vitro study. Virus Res 1999;63:115–9. [31] Jokinen EI, Salo HM, Markkula SE, Immonen AK, Aaltonen TM. Ultraviolet B irradiation modulates the immune system of fish (Rutilus rutilus, Cyprinidae). Part III: lymphocytes. Photochem Photobiol 2001;73:505–12. [32] Siwicki A, Dobosz S, Goryczko K, Kuz´min´ski H, Kohlmann K, Trapkowska S, et al. Cell-mediated immunity in yellow forms of rainbow trout. Pol J Vet Sci 2003;6:49–50. [33] Secombes CJ, van Groningen JJ, Egberts E. Separation of lymphocyte subpopulations in carp Cyprinus carpio L. by monoclonal antibodies: immunohistochemical studies. Immunology 1983;48:165–75. [34] Schneider B, Ambrosius H. Immunohistochemical characterization of lymphocyte populations in carp (Cyprinus carpio L.). Biomed Biochem Acta 1989;48:159–63. [35] Scapigliati G, Romano N, Abelli L, Meloni S, Ficca AG, Buonocore F, et al. Immunopurication of T lymphocytes from sea bass Dicentrarchus labrax (L.). Fish Shellfish Immunol 2000;10:329–41. [36] Scapigliati G, Meloni S, Buonocore F, Bossu` P, Prugnoli D, Secombes CJ. Immunopurication of B lymphocytes from sea bass Dicentrarchus labrax (L.). Mar Biotechnol (NY) 2003;5:214–21.
E. Saito et al. / Fish & Shellfish Immunology 27 (2009) 549–555 [37] Miller SB. Prostaglandins in health and disease: an overview. Semin Arthritis Rheum 2006;36:37–49. [38] Smith WL. The eicosanoids and their biochemical mechanisms of action. Biochem J 1989;259:315–24. [39] Sugimoto Y, Narumiya S, Ichikawa A. Distribution and function of prostanoid receptor: studies from knockout mice. Prog Lipid Res 2000;39: 289–314. [40] Davis P, Baily PJ, Goldenberg MM, Ford-Hutchinson AW. The role of arachidonic acid oxygenation products in pain and inflammation. Annu Rev Immunol 1984;2:335–57. [41] Park JY, Pillinger MH, Abramson SB. Prostaglandin E2 synthesis and secretion: the role of PGE2 synthases. Clin Immunol 2006;119:229–40. [42] Goodwin JS, Bankhurst AD, Messner RP. Suppression of human T-cell mitogenesis by prostaglandin. J Exp Med 1977;146:1719–34. [43] Minakuchi R, Wacholtz MC, Davis LS, Lipsky PE. Delineation of the mechanism of inhibition of human T cell activation by PGE2. J Immunol 1990;145:2616–25. [44] Chouaib S, Welte K, Mertelsmann R, Dupont B. Prostaglandin E2 acts at two distinct pathways of T lymphocyte activation: inhibition of interleukin 2 production and down-regulation of transferring receptor expression. J Immunol 1985;135(2):1172–9. [45] Nataraj C, Thomas DW, Tilley SL, Nguyen M, Mannon R, Koller BH, et al. Receptors for prostaglandin E2 that regulate cellular immune responses in the mouse. J Clin Invest 2001;108:1229–35. [46] Emond V, Fortier MA, Murphy BD, Lambert RD. Prostaglandin E2 regulates both interleukin-2 and granulocyte-macrophage colony-stimulating factor gene expression in bovine lymphocytes. Biol Reprod 1998;58:143–51.
555
[47] Goodwin JS, Ceuppens J. Regulation of the immune response by prostaglandins. J Clin Immunol 1983;3:295–315. [48] Gordon D, Bray M, Morley J. Control of lymphokine secretion by prostaglandins. Nature 1976;262:401–7. [49] Secombes CJ, Clements K, Ashton I, Rowley AF. The effect of eicosanoids on rainbow trout, Onchorhynchus mykiss, leucocyte proliferation. Vet Immunol Immunopathol 1994;42:367–78. [50] Arosh JA, Banu SK, Chapdelaine P, Fortier MA. Temporal and tissue-specific expression of prostaglandin receptor EP2, EP3, EP4, FP and cyclooxygenase 1 and 2 in uterus and fetal membranes during bovine pregnancy. Endocrinology 2004;145:407–17. [51] Poyser NL. The control of prostaglandin production by the endometrium in relation to luteolysis and menstruation. Prostaglandins Leukot Essent Fatty Acids 1995;53:147–95. [52] Lim H, Paria BC, Das SK, Dinchuk JE, Langenbach R, Trzaskos JM, et al. Multiple female reproductive failures in cyclooxygenase 2-deficient mice. Cell 1997;91:197–208. [53] Patel FA, Challis JR. Prostaglandins and uterine activity. Front Horm Res 2001;27:31–56. [54] Challis JR, Sloboda DM, Alfaidy N, Lye SJ, Gibb W, Patel FA, et al. Prostaglandins and mechanisms of preterm birth. Reproduction 2002;124:1–17. [55] Brandon JM. Leucocyte distribution in the uterus during the preimplantation period of pregnancy a phagocyte recruitment to sites of blastocyst attachment in mice. J Reprod Fertil 1993;98:567–76. [56] Redline RW, Lu CY. Specific defects in the anti-listerial immune response in discrete regions of the murine uterus and placenta account for susceptibility to infection. J Immunol 1988;140:3947–55.
Contents lists available at ScienceDirect
Fish & Shellfish Immunology journal homepage: www.elsevier.com/locate/fsi
Suppression of lymphocyte proliferation by ovarian cavity fluid from the viviparous fish Neoditrema ransonnetii (Perciformes; Embiotocidae) Erina Saito*, Osamu Nakamura, Hiroyuki Yamada, Shigeyuki Tsutsui, Tasuku Watanabe School of Marine Biosciences, Kitasato University, Ofunato, Iwate 022-0101, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 10 December 2008 Received in revised form 13 July 2009 Accepted 14 July 2009 Available online 21 July 2009
As the fetus expresses paternal major histocompatibility complex molecules, viviparous vertebrates require sophisticated mechanisms to modulate maternal immunology to ensure successful pregnancy. We anticipated that ovarian cavity fluid (OCF) is likely to feature significantly in the modulation of ovarian cavity immunology. Consequently, we examined the effects of OCF upon leukocyte function in Neoditrema ransonnetii. OCF did not affect phagocytosis or superoxide production by phagocytes. However, OCF suppressed lymphocyte proliferation induced by ConA almost completely. As OCF contained PGE2 at high levels during late pregnancy, we also investigated the effect of PGE2 upon lymphocyte expansion. PGE2 exhibited negative effects upon lymphocyte mitogenesis in a dosedependent manner (10–1000 ng/ml). PGE2 significantly suppressed lymphocyte proliferation when present at levels equivalent to that seen in OCF (30.2 16.1 w 185.4 107.4 ng/ml). Data indicate that PGE2 is one of the key modulatory molecules of the maternal immune system ensuring successful pregnancy in this viviparous species. Ó 2009 Elsevier Ltd. All rights reserved.
Keywords: Viviparity Ovarian cavity fluid Immunomodulation Lymphocyte proliferation Prostaglandin E2 Pregnancy Surfperch
1. Introduction Viviparity requires highly sophisticated maternal control mechanisms in order to modulate immune responses effectively because the spermatozoa and fetuses express paternal major histocompatibility complex (MHC) antigens. It is essential for the success of internal fertilization and maintenance of pregnancy that local immune reactions are regulated. A number of studies have attempted to elucidate the immuno-modulating mechanisms which underlie pregnancy for humans and other mammalian species, and these studies have been largely successful. Such mechanisms include the restricted expression of MHC molecules on the trophoblast [1–4], presence of multiple immunoregulatory factors (e.g., cytokines, hormones, and prostaglandin E2 (PGE2)) [5–8], and the distinct distribution of immune cells across the maternal–fetal interface [9–15]. In stark contrast with mammalian species, this puzzle has been left almost unexplored in non-mammalian viviparous species, despite the fact that viviparity is widespread among vertebrates, with the exception of Aves and Agnatha. There are over one thousand species of viviparous fish [16]. Fish are the first vertebrate group to have developed MHC and T cell receptors (TCR), and, as * Corresponding author. Tel.: þ81 192 44 1907; fax: þ81 192 44 2125. E-mail address: [email protected] (E. Saito). 1050-4648/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.fsi.2009.07.006
a result, exhibit strict immune-rejection against non-self transplants [17–22]. Viviparous fishes are, therefore, the oldest creatures to have faced conflict between the immune system, internal fertilization and fetal growth. Understanding how fish have resolved this conflict would reveal vital information regarding the evolutionary process of viviparity in vertebrates. Embiotocidae fish, also known as ‘surfperch’, possess some unique characteristics among viviparous teleosts; their gestational period is considerably long (more than 6 months), the fetus has very limited yolk reserves resulting in almost complete dependence on maternally supplied nutrients. The fish do not develop a placenta. Instead, fetuses continuously receive maternal secretions, known as the ovarian cavity fluid (OCF). Neoditrema ransonnetii is a common Embiotocidae fish found in Japan. Copulating behavior is observed from September to December, in Iwate, Northern Japan. Internal fertilization appears to occur from late December through to early January. Fetuses retained in the ovarian cavity ingest and vigorously absorb OCF via their hypertrophied hindgut [23]. From late July to early August, newborn are delivered. Through microscopic observation, we previously found that a large number of leukocytes exist in the ovarian cavity [24]. This indicates that inseminated sperm cells and developing fetuses are not immunologically sequestered from the maternal immune system in surfperch. Vigorous phagocytosis of the inseminated sperm cells by macrophages is evident in the ovarian lumen after coitus, and even
550
E. Saito et al. / Fish & Shellfish Immunology 27 (2009) 549–555
during the gestation period [24]. Nevertheless, neither an apparent immune response nor a semi-allogenic rejection appears to occur. Throughout the reproductive cycle, more than 90% of observed leukocytes in the ovarian lumen are monocytes/macrophages, while lymphocytes invariably represent only a small population. These facts suggest the presence of local mechanisms suppressing lymphocyte influx or proliferation. We therefore expected the OCF to play key roles in the immune modulation of the ovarian cavity in order to maintain pregnancy in surfperch. In the current study, we examined the immuno-modulating effect of OCF upon phagocytic functions and lymphocyte proliferation. In addition, we also studied the effect of PGE2, a well-known inflammatory mediator and immuno-modulating agent, because we found that OCF contained high levels of PGE2. 2. Materials and methods 2.1. Fish Fish were collected by angling at Okirai Bay, Iwate, Japan, and were maintained under natural conditions with running sea water in fiber-reinforced plastic tanks at Kitasato University. Fish were fed a commercial diet for marine fish (Fuji-Seifun Co., Japan). To avoid any possible influences that can be induced by pregnancy-related factors, the cells used in all experiments were obtained from the males. 2.2. Collection of plasma and OCF Fishes were anesthetized with 2-phenoxyethanol, and blood collected using heparinized syringes with a 26G needle. Following centrifugation at 700 g for 15 min at 4 C, the supernatants were harvested and stored at 80 C to await analysis. Following laparotomy, the ovaries were dissected and the OCF collected by a micropipette. Following centrifugation, the supernatants were preserved at 80 C and sterilized with a syringe filter unit (pore size, 0.20 mm; ADVANTEC, Japan) before use. 2.3. Cell preparation Blood was drawn aseptically from the caudal vein of anesthetized fish into a syringe containing 1000 IU/ml (the final dilution: 125 IU/ml) of heparin and immediately mixed with 5 volumes of 1 Dulbecco’s phosphate-buffered saline (D-PBS) (Gibco, USA). The diluted blood was then layered on 45% PercollÔ (GE Healthcare, USA) prior to centrifugation at 400 g for 40 min at 20 C. The peripheral blood leukocytes (PBL) layer at the interface was harvested and washed 3 times before use in the following experiments. Following laparotomy, the head kidney was removed and pushed through a stainless mesh in D-PBS. The resulting cell suspension was placed on 35/51% PercollÔ and centrifuged. The macrophage-enriched cells from the 35/51% PercollÔ interface were recovered, washed with 0.01 M PBS, and resuspended in the RPMI 1640 medium (Nissui Pharmaceutical, Japan). These cells are referred to as head kidney leukocytes (HKLs) hereafter. 2.4. Phagocytosis assay HKLs (2 106 cells) in RPMI 1640 were allowed to adhere to a culture coverglass (Matsunami Glass, Japan) for 24 h at 15 C, and non-adherent cells were removed by washing with 0.01 M PBS. Cells were then maintained with 150 ml of OCF or RPMI 1640 at 15 C for 6 h. Following washing, latex beads (L.B, 0.75 mm, Polysciences Inc, USA), opsonized with N. ransonnetii serum, were
added to the well along with the RPMI 1640 medium. After 1 h of incubation at 15 C and washing, the cells on the cover slip were fixed with methanol and stained with 40 -6-diamino-2-phenylindole. The number of phagocytic cells that ingested at least one particle was counted microscopically and the phagocytic activity was expressed as a percentage of the total cell counts. Opsonic activity of OCF was also examined. L.B. were treated with OCF, serum, or RPMI 1640 medium for 1 h at room temperature. Following washing, these were resuspended in RPMI 1640 medium, and added to HKL monolayers on the cover slip. Phagocytic activity was measured as described above.
2.5. Intracellular production of superoxide anions in the macrophages Superoxide anion production by HKLs was determined according to the method described by Chung and Secombes [25]. In brief, the HKLs (1 106 cells/well) were plated onto 96-well microtiter plates (AGC Techno glass, Japan) and allowed to adhere for 24 h at 15 C. After washing with RPMI 1640, the macrophage monolayers were treated with 80 ml of OCF, male serum, or RPMI 1640 for 24 h. After another round of washing, the cells were incubated with 100 ml of NBT solution (1 mg/ml) in RPMI 1640 with or without Zymosan A (2.5 107 particles/well) (Sigma, USA) opsonized with N. ransonnetii serum. After 1 h, the NBT solution was removed and the cells were fixed with methanol. The reduced formazan within the macrophages was solubilized in KOH/DMSO for 5 min and the optical density was measured by a microplate reader (Model 550; BioRad, USA) at 620 nm. 2.6. Cell proliferation assay Isolated PBLs were resuspended in OCF or Leibovitz’s L-15 medium (Gibco) containing 100 IU/ml penicillin and 100 mg/ml streptomycin. The cell suspensions were incubated in 96-well microplates with or without 1 mg/well of concanavalin A (ConA, Sigma) for 72 h at 15 C. Twenty-four hours before the end of cultivation, 5-bromo-20 -deoxyuridine (BrdU, final concentration: 10 mM) was added to the culture. BrdU uptake into nuclear DNA was measured using the Cell Proliferation ELISA kit (GE Healthcare) according to the manufacturer’s instructions.
2.7. Characterization of the proliferating cells To characterize the cells proliferating in response to ConA, we centrifuged and smeared a few of these cells onto glass slides. Following acetone fixation and blocking with 1% FBS, the cells were incubated with anti-IgM rabbit antiserum ( 10,000) [26] for 2 h at room temperature. After washing, the cells were reacted with antirabbit IgG goat IgG conjugated with Alexa Fluor 633 (Funakoshi, Japan) for 1 h. After washing, DNA was fragmented using DNase (Promega, USA) for 1 h at 37 C. The cells were washed again and then exposed to 10 mg/ml of FITC-labeled anti-BrdU mouse mAb (Biomeda Corporation, USA) for 2 h. Lectin staining with ConA was also performed in order to identify the ConA-binding cells. PBLs were fixed with 4% paraformaldehyde and smeared by cytospin. After blocking of endogenous biotin with the avidin/biotin blocking kit (Vector Laboratories, USA), cells were treated with 5 mg/ml of biotin-labeled ConA (Sigma) for 1 h and then incubated with 100 mg/ml of FITC-conjugated streptavidine (Sigma) for 30 min. All fluorescent signals were captured by a confocal laser-scanning microscope (LSM-510META; Carl Zeiss, Germany).
E. Saito et al. / Fish & Shellfish Immunology 27 (2009) 549–555
2.8. Hemagglutination assay To confirm whether OCF interrupted the activity of the mitogen, we examined the hemagglutination activity of ConA in the presence of OCF. First, the lectin activity of OCF was assessed. OCF itself clumped rabbit red blood cells (RRBCs), and this activity was inhibited by 0.2 M L-()-rhamnose or L-()-fucose. Therefore, 0.2 M rhamnose was added to OCF or the L-15 medium and pre-incubated for 1 h at room temperature. ConA (40 mg/ml) was serially diluted with these solutions on 96-well V-bottom microtiter plates (Kartell, Italy) and mixed with the RRBC suspension. 2.9. Effect of PGE2 on PBL proliferation induced by ConA PGE2 concentrations in the plasma and OCF were measured using a commercial ELISA kit (Cayman Chemicals, USA) according to the manufacturer’s instructions. The antibody in this kit shows crossreactivity with PGE3 (43%) and PGE1 (20%). To confirm whether the substances measured in the ELISA was PGE2 or other E series PGs, a low molecular mass fraction of OCF was prepared by ultrafiltration using Vivaspin 5 k (Sartorius Stedim Biotech, Germany) and analyzed by MALDI TOF-MS (Shimadzu biotech, Japan). PBLs were suspended in L-15 medium and plated onto 96-well plates as described above. PGE2 solutions (0.01–1000 ng/ml) containing 0.01% ethanol and 20 mg/ml of ConA were added to each well (50 ml/well). Control cultures were tested with regard to the effects of ethanol at the same final dilution (0.01%). After 72 h of incubation, cell proliferation was determined by the BrdU assay as described earlier. 2.10. Statistical analysis All numerical data are presented as mean SD. Data from the phagocytosis assay and quantification of PGE2 were analyzed by paired and unpaired t tests, respectively. Other data were analyzed using repeated measures ANOVA followed by Bonferroni pairwise comparisons or Dunnett’s multiple comparison test. Differences were considered as statistically significant at the 95% confidence level (p < 0.05). 3. Result 3.1. Effect of OCF on phagocytic functions The phagocytic activity of renal macrophages against microspheres that had been opsonized by serum was not affected by the presence of OCF (Fig. 1). The phagocytic rate of HKLs in the OCF did
Fig. 1. Effect of OCF upon phagocytosis. After HKLs were treated with OCF or RPMI 1640, phagocytic activity was assessed. Data are expressed as mean SD (n ¼ 3) of the percentage of cells phagocytizing L.B.
551
not differ from that in the medium. On the other hand, OCF exhibited significant opsonic activity for HKLs that was nearly equal to that of the serum (Fig. 2). On stimulation with opsonized zymosan, HKLs in the OCF produced superoxide anions at levels almost equal to that in the male serum, but higher than that observed in the medium (Fig. 3). 3.2. Effect of OCF on lymphocyte proliferation In L-15 medium, PBLs underwent proliferation upon stimulation with ConA within 72 h of incubation. This induced proliferation was almost completely abolished in the OCF (Fig. 4). Although ConA is known to act as a T cell mitogen in mammals, the type of cells that respond to ConA in fish has yet to be clarified. Thus, we morphologically characterized cells incorporating BrdU or bearing binding sites for ConA. As shown in Fig. 5A and B , both these cells displayed the typical features of small-sized lymphocytes with a large nucleo-cytoplasmic ratio. When the cells were double stained with anti-BrdU and anti-IgM antibodies, nearly half of the cells incorporating BrdU were IgM negative (Fig. 5A), but the other half was IgM positive (Fig. 5B). Negative controlled cells that were treated without anti-BrdU mAb and anti-IgM antiserum or biotin-labeled ConA were not stained (data not shown). These results indicate that the cells undergoing proliferation upon ConA stimulation were lymphocytes-like cells. To ascertain that the suppression of mitogenesis was a direct result of the inhibitory effects of OCF upon lymphocytes but not from its interference with the mitogen, we further investigated the agglutination activity of ConA against RRBCs in the presence of OCF. Having performed three times of similar experiments, the results showed that the agglutination titer of ConA in the OCF was not significantly different from that in the medium (Table 1). 3.3. PGE2 is the proliferation inhibitory factor in the ovarian cavity PGE2 concentration was measured in plasma and OCF from females. The levels in the plasma of pregnant females in July, i.e., a few weeks before parturition, tended to increase but not significantly. In addition, there was little difference between pregnant and non-pregnant female in plasma PGE2 levels. On the other hand, levels in the OCF in April were 185.4 107.4 ng/ml. This was significantly higher than those of other periods, although the values in April markedly varied between individuals. In July, the PGE2 concentration in the OCF was 30.2 16.1 ng/ml, approximately 10fold higher than in the plasma (Fig. 6). Because the antibody in this
Fig. 2. Opsonic activity of OCF. Phagocytes were brought into contact with L.B. pretreated with OCF, serum, or the RPMI 1640 medium, and incubated for 1 h. Data are expressed as mean SD (n ¼ 5) of the percentage of cells phagocytizing L.B. *, significant at p < 0.05; **, significant at p < 0.01.
552
E. Saito et al. / Fish & Shellfish Immunology 27 (2009) 549–555
Fig. 3. Effect of OCF on superoxide anion production. After the macrophages were incubated in OCF, male serum, or RPMI 1640 medium, intracellular superoxide anions were measured by NBT reduction. Data are expressed as mean SD (n ¼ 5). *, significant at p < 0.05.
kit has cross-reactivity with other E series PGs such as PGE1 and PGE3, OCF was analyzed by MALDI TOF-MS to confirm whether the substances detected in this study were PGE2 or other E series PGs. A peak was detected at almost the same molecular mass as PGE2 (352.5) (Fig. 7). However, peaks corresponding to the molecular masses of PGE1 (354.5) and PGE3 (350.5) were not detected. Therefore, it was concluded that the molecules measured by the kit in this study were PGE2, at least the case of the OCF in July. It was not possible to analyze OCF before July because its amount was too small. We then examined the effect of PGE2 upon lymphocyte proliferation. As shown in Fig. 8, PGE2 strongly inhibited the ConAinduced proliferation of PBLs. This effect was dose-dependent and significant suppression was observed at concentrations from 10 to 1000 ng/ml. 4. Discussion In the current study, we demonstrate that the ovarian cavity fluid (a histotrophe secreted by ovarian tissues) from surfperch exerted a potent inhibitory effect upon lymphocyte-like cells proliferation induced by ConA (Fig. 4) but did not suppress phagocytic functions. Furthermore, we demonstrated that the suppressive activity of the OCF upon those cells was ascribed to, at
Fig. 5. Immunofluorescent detection of proliferating cells and ConA-binding cells. Cells incorporating BrdU were stained with anti-IgM antibody and FITC-labeled anti-BrdU Ab (A). A1 is BrdU (green) single positive cells. A2 is IgM (purple) and BrdU double positive cells. ConA-binding cells were detected by biotin-labeled ConA and FITCconjugated streptavidine (B). Scale bar ¼ 5 mm.
least in part, PGE2 that had accumulated in the OCF in high levels (Fig. 8). ConA acts as a mitogen for mammalian T cells and has shown mitogenic activity for lymphocytes in several species of fish [27–32]. Though lymphocyte cell markers have been elucidated in some fish species, such as carp and sea bass [33–36], it is still difficult to distinguish T cells, B cells, and NK cells in the majority of fish species, including N. ransonnetii, because of unidentified those markers. Accordingly, it is unclear which types of lymphocytes are capable of responding to ConA stimulation in fish. The present study tried to identify the properties of the proliferating cells using double staining with anti-IgM antiserum and an anti-BrdU antibody. It was found that two types of cells, which were IgMþ and IgM, existed in the cells incorporating BrdU (Fig. 5). However, it was doubtful whether this anti-IgM does not react with T cells because about 50% of thymocytes can be stained using this antibody (data not shown). Therefore, the specific cell types could not be identified in the current study. However, they showed typical lymphocyte-like morphology, and at least half of them were IgM negative. Irrespective of which type of cells these are, it is important to elucidate which specific types of cells (Th, Tc, or others) can be suppressed by PGE2. Consequently, characterization of these cells will be the main focus in our future research efforts.
Table 1 The effected of OCF on agglutination of RRBC by ConA. ConA concentration (mg/ml) Fig. 4. Effect of OCF on PBL proliferation. The isolated PBLs were resuspended in OCF or L-15, with or without 1 mg/well of ConA. After 72 h, cell proliferation was measured on the basis of BrdU uptake into nuclear DNA. Data are expressed as mean SD (n ¼ 5). *, significant at p < 0.05.
OCF L-15
20
10
5
2.5
1.25
0
þ þ
þ þ
þ, agglutination; , no agglutination (n ¼ 3).
E. Saito et al. / Fish & Shellfish Immunology 27 (2009) 549–555
553
Fig. 8. Effect of PGE2 on PBL proliferation. PGE2 added to PBL at a dose of 0.01– 1000 ng/ml with 20 mg/ml ConA. After 72 h of incubation, cell proliferation was determined. Data are expressed as mean SD (n ¼ 5). *, significant at p < 0.05; **, significant at p < 0.01.
Fig. 6. PGE2 concentrations in the plasma and OCF during late pregnancy. A is the levels in plasma and B is in OCF. PGE2 concentrations were measured by ELISA. Data are expressed as mean SD (n ¼ 5). *, significant at p < 0.05.
It may be possible to explain the results shown by Fig. 4 in an alternative manner; OCF may inhibit mitogenesis not by suppressing the responsiveness of lymphocyte-like cells but rather, by interfering with the activity of the mitogen. To examine this possibility, we tested the agglutination activity of ConA against RRBCs in the OCF. As shown in Table 1, the agglutination titer of ConA in the OCF was not reduced compared with that of the medium. Therefore, we concluded that OCF could depress the proliferating ability of ConA-responding cells without interrupting the mitogen. This property of OCF may partially explain why the
Fig. 7. Characterization of molecules included in OCF by MALDI TOF-MS analysis. Low molecular fraction of OCF in July were analyzed by MALDI TOF-MS.
number of lymphocytes is not increased under the existence of immunologically foreign sperm cells and the embryos in the ovarian cavity. PGE2, an arachidonic acid derivative, is a potent bioactive substance involved in a diverse range of physiological and pathological phenomena, including salt and water excretion, bone resorption, gastric acid secretion, fever, mediation of pain, and the stimulation of inflammatory responses [37–41]. In mammals, PGE2 has been known to possess immunosuppressive activity as well; it inhibits T cell proliferation, lymphokine production, and cytotoxicity [42–48]. Although the inflammatory and immunosuppressive functions of PGE2 have not been well elucidated in fish, Secombes et al. reported that PGE2 inhibited mitogenesis in rainbow trout leukocytes in response to PHA-P [49]. In this study, it was found that OCF contained PGE2 in much higher concentrations than in plasma (Fig. 6). To ascertain whether other E series PGs were also present, OCF was analyzed by MALDI TOF-MS. As a result, a peak was detected at the appropriate position for PGE2, but neither PGE1 nor PGE3 were detected (Fig. 7). These results indicated that PGE2 exerts a significant impact upon the maintenance of gestation in N. ransonnetii. Consequently, we demonstrated that PGE2 inhibited lymphocyte proliferation in a dose-dependent manner; PGE2 significantly suppressed lymphocytic mitogenesis at concentrations from 10 to 1000 ng/ml. The mean concentration of PGE2 in the OCF in gravid females was much higher than that in the plasma in pregnant and non-pregnant females. It was not possible to determine PGE2 levels in the OCF of non-pregnant fish, because the volume of the fluid was too small to collect. These results indicated that PGE2 plays a role as an immuno-suppressor in the ovarian cavity during pregnancy. In the current study, we used OCF only from April to July. This was because it was difficult to collect sufficient amounts of OCF from the mother in the early stages of conception. The gestation period of N. ransonnetii is rather long, and the nature of OCF appears to change significantly during this period. OCF appears more dense and proteinaceous during April, the mid stage of gestation [26]. Therefore, PGE2 levels in the OCF may vary with the progression of pregnancy. In mammals, PGE2 is involved in multiple reproductive events such as ovulation, fertilization, implantation, maintenance of pregnancy, and parturition [50–54]. Thus, other functions of
554
E. Saito et al. / Fish & Shellfish Immunology 27 (2009) 549–555
PGE2 in surfperch reproduction should also be considered, and decreased levels of PGE2 in the OCF during later stages may be a transient phenomenon, possibly associated with parturition. Further research should investigate the level of PGE2 in OCF during other stages of gestation to understand the roles of this multifunctional mediator. On the other hand, one should assume that other immunomodulating substances, like sex hormones and suppressive cytokines, are also involved in the complex mechanisms leading to successful pregnancy in viviparous fish. Each of these additional molecules may play more significant roles in the immune regulation of earlier stages, if the OCF present during those periods contains reduced amounts of PGE2, as discussed above. In all experiments in the present study, we used leukocytes obtained from the head kidney and peripheral blood of male fish. Female N. ransonnetii exhibits a long mating and gestation period, and the period when females participate in neither copulation nor gestation lasts only around one month. Leukocyte responsiveness may also vary concomitant with female reproductive cycle. Therefore we needed to use males, in order to exclude any possible influences of reproduction-related factors which might have already affected maternal leukocytes. In contrast to its suppressive effect on lymphocytes, OCF did not interfere with phagocytosis or superoxide production by phagocytes. In addition, OCF possessed opsonic activity almost equal to that of the serum. Although opsonic molecules in the OCF have yet to be identified, our former study showed that OCF contained IgM, though levels were much lower than that in the plasma [26]. In addition, lectin, which was suggested to be present in OCF in the current study, may also act as an opsonin. These results indicate that the leukocyte functions associated with phagocytosis are maintained normally in the ovarian cavity of N. ransonnetii even during the gestational period. Because ovarian cavity is separated from the external environment only by a short oviduct, fetuses and ovarian epithelia may be easily threatened by invasion of microorganisms. Therefore, it seems reasonable that phagocytic functions are retained normally for the protection of fetal and maternal tissues, as in the case of mammalian genital tracts in gravid females [13,55,56]. In summary, we have provided evidence herein for the existence of immuno-modulating mechanisms during pregnancy and in particular, one specific candidate molecule, PGE2, in a viviparous fish, N. ransonnetii. Since severe conflicts between the adaptive immune system and viviparity have arisen along with the emergence of viviparous fish, the fact that surfperch exploit PGE2 as an immunomodulator for successful pregnancy would provide a further basis for understanding the evolutionary process of viviparity in vertebrates. Acknowledgements The authors thank Prof. Shunsuke Moriyama (Kitasato University) for his expertise with MALDI TOF-MS. References [1] Sargent IL. Maternal and fetal immune responses during pregnancy. Exp Clin Immunogenet 1993;10:85–102. [2] Murphy SP, Choi JC, Holtz R. Regulation of major histocompatibility complex class II gene expression in trophoblast cells. Reprod Biol Endocrinol 2004;2:52. [3] Bacon SJ, Ellis SA, Antczak DF. Control of expression of major histocompatibility complex genes in horse trophoblast. Biol Reprod 2002;66:1612–20. [4] Carter J. The maternal immunological response during pregnancy. Oxf Rev Reprod Biol 1984;6:47–128. [5] Parhar RS, Kennedy TG, Lala PK. Suppression of lymphocyte alloreactivity by early gestational human decidua. I. Characterization of suppressor cells and suppressor molecules. Cell Immunol 1988;116:392–410.
[6] Lala K, Kennedy TG, Parhar RS. Suppression of lymphocyte alloreactivity by early gestational human decidua. II. Characterization of suppressor mechanisms. Cell Immunol 1988;116:411–22. [7] Piccinni MP. T-cell cytokines in pregnancy. AJRI 2002;47:289–94. [8] Jones RL, Hannan NJ, Kaitu’u TJ, Zhang J, Salamonsen LA. Identification of chemokines important for leukocyte recruitment to the human endometrium at the times of embryo implantation and menstruation. J Clin Endocrinol Metab 2004;89:6155–67. [9] Lawrence R, Church JA, Richards W, Borzy M. Immunological mechanisms in the maintenance of pregnancy. Ann Allergy 1980;44:166–73. [10] Vince GS, Johnson PM. Leucocyte populations and cytokine regulation in human uteroplacental tissues. Biochem Soc 2000;28:191–5. [11] Ku¨hnert M, Strohmeier R, Stegmu¨ller M, Halberstadt E. Changes in lymphocyte subsets during normal pregnancy. Eur J Obstet Gynecol Reprod Biol 1998;76:147–51. [12] Trundley A, Moffett A. Human uterine leukocytes and pregnancy. Tissue Antigens 2004;63:1–12. [13] Hunt JS. Immunologically relevant cells in the uterus. Biol Reprod 1994;50:461–6. [14] Milne SA, Henderson TA, Kelly RW, Saunders PT, Baird DT, Critchley HO. Leukocyte populations and steroid receptor expression in human firsttrimester decidua; regulation by antiprogestin and prostaglandin E analog. J Clin Endocrinol Metab 2005;90:4315–21. [15] Ozenci CC, Korgun ET, Demir R. Immunohistochemical detection of CD45þ, CD56þ, and CD14þ cells in human decidua during early pregnancy. Early Pregnancy 2001;5:164–75. [16] Wourms JP. Viviparity: the maternal-fetal relationship in fishes. Am Zool 1981;21:473–515. [17] McKinney EC, McLeod TF, Sigel MM. Allograft rejection in a holostean fish, Lapososteus Platyrhincus. Dev Comp Immunol 1981;5:65–74. [18] Kaastrup P, Stet RJ, Tigchelaar AJ, Egberts E, van Muiswikel WB. A major histocompatibility locus in fish: serological identification and segregation of transplantation antigens in the common carp (Cyprinus carpio, L.). Immunogenetics 1989;30:284–90. [19] Nagelar JJ, Cloud JG, Wheeler PA, Thorgaard GH. Testis transplantation in male rainbow trout (Onchorynchus mykiss). Biol Reprod 2001;64:644–6. [20] Cardwell TN, Sheffer RJ, Hedrick PW. MHC variation and tissue transplantation in fish. J Hered 2001 Jul-Aug;92:305–8. [21] Wright Jr JR, Pohajdak B, Xu BY, Leventhal JR. Piscine islet xenotransplantation. ILAR J 2004;45:314–23. [22] Romano N, Baldassini MR, Buonocore F, Picchietti S, Mastrolia L, Abelli L. In vivo allograft rejection in a bony fish Dicentrarchus labrax (L): characterization of effector lymphocytes. Cell Tissue Res 2005 Sep;321:353–63. [23] Nakamura O, Tazumi Y, Muro T, Yasuhara Y, Watanabe T. Active uptake and transport of protein by the intestinal epithelial cells in embryo of viviparous fish, Neoditrema ransonnetii (Perciformes: Embiotocidae). J Exp Zool 2004;301A:38–48. [24] Tazumi Y, Nakamura O, Watanabe T. Intraovarioan cavity leucocytes of viviparous fish, Neoditrema ransonnetii (Perciformes, Embiotocidae). Zool Sci 2004;21:739–46. [25] Chung S, Secombes CJ. Analysis of events occurring within teleost macrophages during the respiratory burst. Comp Biochem Physiol 1988;89B:539–44. [26] Nakamura O, Kudo R, Aoki H, Watanabe T. IgM secretion and absorption in the materno-fetal interface of a viviparous teleost, Neoditrema ransonnetii (Perciformes; Embiotocidae). Dev Comp Immunol 2006;30:493–502. [27] Caspi RR, Shahrabani R, Kehati-Dan T, Avtalion RR. Heterogeneity of mitogenresponsive lymphocytes in carp (Cyprinus carpio). Dev Comp Immunol 1984;8:61–70. [28] Le Morvan-Rocher C, Troutaud D, Deschaux P. Effects of temperature on carp leukocyte mitogen-induced proliferation and nonspecific cytotoxic activity. Dev Comp Immunol 1995;19:87–95. [29] Van der Heijden MH, Booms GH, Tanck MW, Rombout JH, Boon JH. Influence of flumequine on in vivo mitogen responses on European eel (Anguilla anguilla L., 1758) lymphoid cells. Vet Immunol Immunopathol 1995;47:143–52. [30] Siwicki AK, Pozet F, Morand M, Volatier C, Terech-Majewska E. Effects of iridovirus-like agent on the cell-mediated immunity in sheatfish (Silurus glanis)dan in vitro study. Virus Res 1999;63:115–9. [31] Jokinen EI, Salo HM, Markkula SE, Immonen AK, Aaltonen TM. Ultraviolet B irradiation modulates the immune system of fish (Rutilus rutilus, Cyprinidae). Part III: lymphocytes. Photochem Photobiol 2001;73:505–12. [32] Siwicki A, Dobosz S, Goryczko K, Kuz´min´ski H, Kohlmann K, Trapkowska S, et al. Cell-mediated immunity in yellow forms of rainbow trout. Pol J Vet Sci 2003;6:49–50. [33] Secombes CJ, van Groningen JJ, Egberts E. Separation of lymphocyte subpopulations in carp Cyprinus carpio L. by monoclonal antibodies: immunohistochemical studies. Immunology 1983;48:165–75. [34] Schneider B, Ambrosius H. Immunohistochemical characterization of lymphocyte populations in carp (Cyprinus carpio L.). Biomed Biochem Acta 1989;48:159–63. [35] Scapigliati G, Romano N, Abelli L, Meloni S, Ficca AG, Buonocore F, et al. Immunopurication of T lymphocytes from sea bass Dicentrarchus labrax (L.). Fish Shellfish Immunol 2000;10:329–41. [36] Scapigliati G, Meloni S, Buonocore F, Bossu` P, Prugnoli D, Secombes CJ. Immunopurication of B lymphocytes from sea bass Dicentrarchus labrax (L.). Mar Biotechnol (NY) 2003;5:214–21.
E. Saito et al. / Fish & Shellfish Immunology 27 (2009) 549–555 [37] Miller SB. Prostaglandins in health and disease: an overview. Semin Arthritis Rheum 2006;36:37–49. [38] Smith WL. The eicosanoids and their biochemical mechanisms of action. Biochem J 1989;259:315–24. [39] Sugimoto Y, Narumiya S, Ichikawa A. Distribution and function of prostanoid receptor: studies from knockout mice. Prog Lipid Res 2000;39: 289–314. [40] Davis P, Baily PJ, Goldenberg MM, Ford-Hutchinson AW. The role of arachidonic acid oxygenation products in pain and inflammation. Annu Rev Immunol 1984;2:335–57. [41] Park JY, Pillinger MH, Abramson SB. Prostaglandin E2 synthesis and secretion: the role of PGE2 synthases. Clin Immunol 2006;119:229–40. [42] Goodwin JS, Bankhurst AD, Messner RP. Suppression of human T-cell mitogenesis by prostaglandin. J Exp Med 1977;146:1719–34. [43] Minakuchi R, Wacholtz MC, Davis LS, Lipsky PE. Delineation of the mechanism of inhibition of human T cell activation by PGE2. J Immunol 1990;145:2616–25. [44] Chouaib S, Welte K, Mertelsmann R, Dupont B. Prostaglandin E2 acts at two distinct pathways of T lymphocyte activation: inhibition of interleukin 2 production and down-regulation of transferring receptor expression. J Immunol 1985;135(2):1172–9. [45] Nataraj C, Thomas DW, Tilley SL, Nguyen M, Mannon R, Koller BH, et al. Receptors for prostaglandin E2 that regulate cellular immune responses in the mouse. J Clin Invest 2001;108:1229–35. [46] Emond V, Fortier MA, Murphy BD, Lambert RD. Prostaglandin E2 regulates both interleukin-2 and granulocyte-macrophage colony-stimulating factor gene expression in bovine lymphocytes. Biol Reprod 1998;58:143–51.
555
[47] Goodwin JS, Ceuppens J. Regulation of the immune response by prostaglandins. J Clin Immunol 1983;3:295–315. [48] Gordon D, Bray M, Morley J. Control of lymphokine secretion by prostaglandins. Nature 1976;262:401–7. [49] Secombes CJ, Clements K, Ashton I, Rowley AF. The effect of eicosanoids on rainbow trout, Onchorhynchus mykiss, leucocyte proliferation. Vet Immunol Immunopathol 1994;42:367–78. [50] Arosh JA, Banu SK, Chapdelaine P, Fortier MA. Temporal and tissue-specific expression of prostaglandin receptor EP2, EP3, EP4, FP and cyclooxygenase 1 and 2 in uterus and fetal membranes during bovine pregnancy. Endocrinology 2004;145:407–17. [51] Poyser NL. The control of prostaglandin production by the endometrium in relation to luteolysis and menstruation. Prostaglandins Leukot Essent Fatty Acids 1995;53:147–95. [52] Lim H, Paria BC, Das SK, Dinchuk JE, Langenbach R, Trzaskos JM, et al. Multiple female reproductive failures in cyclooxygenase 2-deficient mice. Cell 1997;91:197–208. [53] Patel FA, Challis JR. Prostaglandins and uterine activity. Front Horm Res 2001;27:31–56. [54] Challis JR, Sloboda DM, Alfaidy N, Lye SJ, Gibb W, Patel FA, et al. Prostaglandins and mechanisms of preterm birth. Reproduction 2002;124:1–17. [55] Brandon JM. Leucocyte distribution in the uterus during the preimplantation period of pregnancy a phagocyte recruitment to sites of blastocyst attachment in mice. J Reprod Fertil 1993;98:567–76. [56] Redline RW, Lu CY. Specific defects in the anti-listerial immune response in discrete regions of the murine uterus and placenta account for susceptibility to infection. J Immunol 1988;140:3947–55.