Immunohistological Characterization of Macrophage Migration Inhibitory Factor Expression in Plasmodium falciparum-Infected Placentas
http://www.100md.com
感染与免疫杂志 2005年第6期
Division of Parasitic Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Public Health Service, U.S. Department of Health and Human Services, Atlanta, Georgia 30333
Center for Tropical and Emerging Global Diseases and Department of Infectious Diseases, College of Veterinary Medicine, University of Georgia, Athens, Georgia 30602
Department of Pathology, Egleston Children's Hospital, Emory University School of Medicine, Atlanta, Georgia 30322
Vector Biology and Control Research Center, Kenya Medical Research Institute, Kisumu, Kenya
Atlanta Research and Education Foundation, Atlanta, Georgia 30033
Department of Microbiology, Faculty of Science, Mahidol University, Bangkok 10400, Thailand
Roll Back Malaria, World Health Organization, Geneva, Switzerland
ABSTRACT
Previously, we have shown that macrophage migration inhibitory factor (MIF) was highly elevated in the placental intervillous blood (IVB) of Plasmodium falciparum-infected women. Here, we compared the expression of MIF in placental tissues obtained from P. falciparum-infected and -uninfected women. Immunoperoxidase staining showed a consistent pattern of MIF expression in syncytiotrophoblasts, extravillous trophoblasts, IVB mononuclear cells, and amniotic epithelial cells, irrespective of their malaria infection status. Cytotrophoblast, villous stroma, and Hofbauer cells showed focal staining. Only amniotic epithelial and IVB mononuclear cells from P. falciparum-infected placentas exhibited significantly higher level of MIF expression than uninfected placentas. Stimulation of syncytilized human trophoblast BeWo cells with P. falciparum-infected erythrocytes that were selected to bind these cells resulted in significant increases in MIF secretion, whereas control erythrocytes, lipopolysaccharides, and synthetic -hematin had minimal effect. These findings suggest that placental malaria modulates MIF expression in different placental compartments.
INTRODUCTION
Pregnant women are at a greater risk of acquiring Plasmodium falciparum malaria infection than nonpregnant women (14). In infected women, P. falciparum parasites accumulate in the placenta, leading to placental malaria (PM). Although both humoral and cellular immune factors have been implicated in immunity against PM (6, 11, 15-17), it remains a paradox how the immune system is able to mediate its protective effect in the presence of immunosuppressive hormones, such as glucocorticoids, which are elevated during pregnancy (18, 19).
Recently, we proposed that macrophage migration inhibitory factor (MIF), which is the only cytokine thus far known to counterregulate the immunosuppressive effect of glucocorticoids (8), may play an important role in modulating protective immune responses to infectious agents such as malaria parasites at the placental level (9). MIF is a unique cytokine in that it has properties of a cytokine, enzyme, and hormone (7). MIF was originally identified as a factor produced by T cells capable of inhibiting the random migration of mononuclear cells in vitro and is now known to be produced by several different cell types in various organs, including macrophages, lymphocytes, the pituitary, fibroblasts, and the placenta (7). MIF is released upon the activation of macrophages by various proinflammatory stimuli such as lipopolysaccharide (LPS), toxic shock syndrome toxin 1, malaria parasites, tumor necrosis factor alpha, and gamma interferon (7, 13). It is also involved in the activation of macrophages and killing of intracellular parasites (7) and in regulating natural killer cell activity (1).
Our studies demonstrated that MIF is elevated 300 to 500 fold in placental intervillous blood (IVB) plasma compared to peripheral blood plasma (9). In PM-positive women, there was a significant increase in IVB MIF levels compared to PM-negative women. We also found that placental intervillous blood mononuclear cells produced higher levels of MIF than peripheral blood mononuclear cells. Based on these observations, we have proposed that increased levels of MIF in the intervillous blood could play an important role in the activation of macrophages by overcoming the immunosuppressive effect of corticoid hormones in the placenta and thus eventually helping in the clearance of malaria parasites.
The presence of MIF protein and MIF mRNA in first-trimester human placentas was reported previously. By immunocytochemical methods, MIF was shown to be present in the placental villi, cytotrophoblasts, trophoblast cell islands, endometrium, and decidua of human first-trimester pregnancy (2). A recent study has reported the presence of MIF in the amniotic fluid, with highest levels found in term placentas (12). In the same study, MIF was also found in different cell layers of the extraembryonic membranes (12). However, a systematic investigation to characterize the expression of MIF in various cellular compartments of term placenta has been lacking.
In this study, we attempted to determine the source of MIF in various tissue compartments of term placentas and how its expression is altered in the presence of PM. In addition, we also assessed whether P. falciparum parasites can induce BeWo cells (trophoblastic origin) to secrete MIF.
MATERIALS AND METHODS
Study site, participants, and samples. Placental tissue samples were obtained from a subset of subjects who were enrolled at the New Nyanza Provincial General Hospital, Kisumu, Kenya, to participate in a cohort study to assess the impact of PM on mother-to-child human immunodeficiency virus type 1 transmission in Kenya (5). Kisumu experiences intense perennial transmission of Plasmodium falciparum malaria with two peak transmission periods occurring from November to December and May to July.
Informed consent was obtained from all patients associated with this study. The human subject experimental guidelines of the Institutional Review Board of the Centers for Disease Control and Prevention, the University of Georgia, and the Kenya Medical Research Institute Ethical Review Committee were strictly followed.
Determination of parasitemia status. The malaria infection status was confirmed by histological examination of buffered, formalin-fixed, and hematoxylin-and-eosin-stained placental sections. Placental tissues that showed the presence of malaria parasites by histological examination were categorized as PM positive, and those showed absence of parasites were categorized as the PM-negative group for this study. All the nine placentas in the PM-positive group had evidence of malaria pigment deposits. Among the nine placentas in the PM-negative groups, three had low levels of malaria pigments, indicating a past infection. However, MIF expression levels were not significantly different between the pigment-positive and pigment-negative placentas in the PM-negative group (data not shown).
Selection of placental tissues for immunohistology. Placentas collected from human immunodeficiency virus-seronegative women were selected for this study. We selected placentas from primigravidae, secundigravidae, and multigravidae in a random manner. There was no significant difference in the median age or gravidity between the PM-positive and PM-negative groups.
Collection and processing of placental tissue samples. Within 12 h of placental expulsion, several small pieces (1 by 1 by 0.3 cm) of accentric placental tissue spanning the entire thickness of the placental disk were excised and immediately placed into 10% neutral buffered formalin. Tissues were fixed overnight, then transferred to 80% ethanol, and stored until being embedded in paraffin. Placental tissue sections (3 μm thickness) obtained from sectioning the paraffin-embedded placental sample blocks were mounted onto polylysine-coated slides before MIF staining. A paraffin-embedded tonsil section from a patient with tonsillitis was used as the positive control tissue. The placental sections were stained with hematoxylin and eosin by standard procedures.
MIF immunohistochemistry. Deparaffinized and rehydrated tissue sections were placed in a moisture chamber box and incubated for 2 h in a water bath at 58°C. Antigen retrieval was performed by boiling the sections in Declere buffer (Cell Marque Corporation, Hot Springs, AR) in a stainless steel pressure cooker for 15 min. To suppress endogenous peroxidase activity, the slides were incubated for 30 min with 3% H2O2. After being washed with deionized water, samples were treated with 3% bovine serum albumin and incubated for 1 h at room temperature to block nonspecific binding. Goat anti-human MIF polyclonal antibody (R&D systems Inc, Minneapolis, MN) and purified goat immunoglobulin G isotype control antibody (Southern Biotechnology Associates, Inc., Birmingham, AL) were used at a 1:500 dilution for staining. Tissue sections were incubated with the primary antibodies overnight at 4°C and washed three times with phosphate-buffered saline (PBS). Then, the sections were incubated for 30 min at room temperature with biotinylated rabbit anti-goat antibody from the universal DAKO LSAB+ kit (DAKO Corporation, Carpinteria, CA) and washed three times with PBS. Horseradish peroxidase-labeled streptavidin from the universal DAKO LSAB+ kit was added to the slides, and the slides were incubated for 30 min at room temperature. After being washed with PBS, the slides were incubated with the chromogen substrate solution (3,3'-diaminobenzidine chromogen solution) for 5 min. After being washed with deionized water, the sections were counterstained with hematoxylin and mounted with aqueous mounting medium before being viewed under a light microscope.
Scoring system. An experienced pathologist (C.A.) examined the placental tissues, and all the sections were coded so that the reader would not know the PM infection status of samples. All sections were evaluated for the intensity of MIF staining and assigned a score ranging from 0 to 4, with the following values: 0 (negative), 1 (weak), 2 (moderate), 3 (strong), and 4 (intense strong). The distribution pattern of MIF staining was further characterized as focal (staining in some areas only), segmental (large segments were stained), or diffuse (staining in all the areas). Staining was individually evaluated in each placental compartment, i.e., amniotic epithelium, syncitio- and cytotrophoblasts, villous stromata, Hofbauer cells, and maternal decidua.
In vitro production of MIF by trophoblasts. A human choriocarcinoma cell line, BeWo, known to have characteristics of trophoblast cells, was obtained from the American Type Culture Collection (Manassas, VA). For the in vitro study, BeWo cells were first cultured in 24-well plastic culture plates (Corning) at 3 x 104 cells/well in minimal essential medium (supplemented with 10% fetal bovine serum) and treated with 40 μM of forskolin (Sigma-Aldrich, St. Louis, MO) (20) to induce syncytiotrophoblast formation. Two days later, the culture was washed once with the same culture medium and then stimulated with 3 x 106 parasitized erythrocytes; the parasites remained in the culture for the duration of the experiment. The parasite lines used for these experiments were from the 3D7 laboratory strain preselected to bind syncytialized BeWo cells by several cycles of selection (N. Lucchi et al., unpublished data). The parasites used for stimulation were semisynchronized unpurified cultures containing mainly late-stage trophozoites. These parasites were washed in culture medium twice before being used. Normal human erythrocytes (red blood cells) that were used to cultivate P. falciparum parasites were used as the control. In addition, synthetic hematin (-hematin prepared from hemin by the modified method) (100 μg/ml; Sigma-Aldrich, St. Louis, MO) (10) and 10 μg/ml of LPS (Sigma-Aldrich, St. Louis, MO) were also used to stimulate syncytialized BeWo cells. Supernatants from the cultures were collected at 0, 4, 8, 12, 24, and 48 h after stimulation and stored at –80°C until being used for measuring MIF levels.
Measurement of MIF levels in supernatants by enzyme-linked immunosorbent assay. A double sandwich enzyme-linked immunosorbent assay was used to measure MIF levels in culture supernatants as previously described (9). MIF was detected with a capture mouse monoclonal anti-human MIF antibody (clone 12302.2) and a biotinylated goat anti-human MIF antibody from R&D Systems, Inc. (Minneapolis, MN). MIF concentrations were determined using a standard curve obtained from the known concentration of cytokine standards included in each assay plate.
Statistical analyses. STATVIEW for Windows (SAS Institute, Inc., version 5) was used for data analysis, as shown in Table 1. Since data were not randomly distributed, the Mann-Whitney test was used. The Student t test was used to compare in vitro MIF data from BeWo cell experiments. P values were considered statistically significant at <0.05.
RESULTS
Immunohistochemical study of MIF. We used nine placental tissue samples from PM-negative women and nine tissue samples from PM-positive women. Immunohistochemical localization of MIF protein with a goat anti-human MIF antibody showed various degrees of positive reactivity for MIF in all placental tissues examined. The results are summarized in Table 1, and the distribution of MIF in various cellular compartments is described below.
MIF in the trophoblast layers of chorionic villi. MIF expression was commonly seen in syncytiotrophoblast layers of villi. Although its expression was widely seen in different segments of the placental sections examined, its distribution was segmental in some placentas, especially in PM-negative placentas, as shown in Fig. 1A. On the contrary, placentas from PM-positive group showed diffuse cytoplasmic staining in the syncytiotrophoblast cells (Fig. 1B). It is important to point out that syncytiotrophoblast layers from the PM-positive group (Fig. 1B) consistently exhibited intense cytoplasmic staining of MIF when compared to the PM-negative group, but the difference was not significant (Table 1).
Villous cytotrophoblast cells showed distinctly positive MIF staining. However, the staining was not uniformly expressed in all the cytotrophoblast layers of chorionic villi, and its distribution was focal. There was no appreciable difference in the pattern or intensity of MIF staining in cytotrophoblastic cells between the PM-negative (Fig. 1C) and PM-positive groups (Table 1).
MIF in the villous compartment of chorionic villi. Villous stromal cells showed MIF expression, especially in primary or stem villi (Fig. 1E). The MIF staining pattern in the villous stromal compartment did not differ between the PM-negative and PM-positive groups. Hofbauer cells (specialized fetal macrophages) in the chorionic villi of the placenta also showed positive staining with MIF (Fig. 1E and 1I). MIF staining in Hofbauer cells was generally more intense in PM-positive than in PM-negative placentas, but the difference was not statistically significant (Table 1). No MIF staining was found in the villous capillaries of chorionic villi (Fig. 1I).
MIF in intervillous blood mononuclear cells. Mononuclear cell and macrophage populations were intensely stained for MIF expression (Fig. 1D). These cell populations were commonly found in the intervillous space of PM-positive placentas. On the other hand, PM-negative placental sections had very few intervillous blood mononuclear and macrophage cell populations. Mononuclear cells in PM-positive placentas showed significantly higher intensity of MIF staining than PM-negative placentas (P < 0.02) (Table 1).
Localization of MIF at the interface of fetal and maternal compartments in placental tissue sections. Extravillous trophoblastic cells, which are fetus derived and concentrated on the maternal side of the placenta, showed an intense cytoplasmic staining of MIF. MIF expression was uniformly distributed in these cells (Fig. 1G). There was no appreciable difference in the intensity and distribution pattern of MIF staining between PM-positive and PM-negative groups.
Maternal decidual cells were generally negative for MIF expression. However, occasionally weak staining of MIF was seen focally in some decidua. Amniotic epithelial cells also uniformly showed intense staining for MIF expression (Fig. 1F). Interestingly, the amniotic epithelial cells were more intensely stained in the PM-positive group (Fig. 1F) than in the PM-negative group (Fig. 1H) (P < 0.03) (Table 1). This difference was the most notable change among various placental compartments examined in this study.
In vitro production of MIF by a choriocarcinoma cell line of trophoblastic origin Since the intensity of MIF staining was prominent in the syncytiotrophoblastic layer of PM-positive placentas compared to PM-negative placentas, we assessed the ability of syncytiotrophoblast-binding P. falciparum to induce the trophoblast cell line BeWo to produce MIF. Supernatants from unstimulated syncytialized BeWo cells showed that these cells constitutively produce MIF. MIF levels increased significantly and progressively for at least 48 h after stimulation with the P. falciparum-infected erythrocytes that were preselected to bind BeWo cells (Fig. 2). In the erythrocyte stimulated control group, secreted MIF levels did not significantly increase compared to baseline levels. Likewise, LPS and synthetic -hematin did not cause any significant increase in MIF levels compared to baseline levels.
We also determined if P.falciparum-infected erythrocytes that were preselected for binding to syncytialized BeWo cells (7XP) induced higher levels of MIF than the unselected 3D7 parental parasite line in three independent experiments. The results from these experiments showed that BeWo cells stimulated for 48 h with unselected 3D7 parasites produced significantly less MIF (arithmetic mean ± standard error, 1,629 ± 215 pg/ml) than the preselected 7XP parasites (2,671 ± 223 pg/ml) (P < 0.01). The background MIF level in the unstimulated (medium only) and normal erythrocyte-stimulated cultures were 1,065 ± 322 pg/ml and 918 ± 405 pg/ml, respectively. There was no significant difference in the MIF level between background control groups and the unselected 3D7 line-stimulated group (P > 0.05). However, the MIF level in the preselected 7XP group was significantly higher than in the background control groups (P < 0.01).
DISCUSSION
In a previous study, we demonstrated that MIF was present at very high levels in the IVB plasma compared to peripheral and cord plasma and that its level was further increased in the presence of placental malaria infection (9). Although we had shown that IVB mononuclear cells were an important source of MIF production, the contribution of other placental compartments, especially in the context of P. falciparum infection, was not known. This study has demonstrated that MIF is expressed in various cellular compartments of the term placenta, such as syncytiotrophoblasts, cytotrophoblasts, extravillous trophoblasts, Hofbauer cells, stromal cells, amniotic epithelial cells, and mononuclear cells in the intervillous space. Comparision between PM-positive and PM-negative placentas revealed significantly higher levels of MIF expression in the amniotic epithelial cells and IVB mononuclear cells. Although there was no significant quantitative difference in the syncytiotrophoblast expression of MIF between PM-infected and PM-uninfected placentas, there were some qualitative differences as seen by the uniform expression of MIF in the PM-positive placentas, while PM-negative placentas often showed segmental expression. The ability of sequestered P. falciparum parasites in the placenta to increase MIF production by syncytiotrophoblast cells was further evident from in vitro experiments, which showed that the trophoblast cell line, BeWo, specifically secreted MIF in response to cytoadherence of P. falciparum-infected erythrocytes but not other nonspecific stimuli. These observations suggest that IVB momonuclear cells and syncytiotrophoblasts could be important sources of MIF production in the placental IVB.
Amniotic epithelial cells widely showed the expression of MIF, with the PM-positive placentas showing the highest levels of MIF expression. We do not have any clear explanation of how amniotic cells that do not come in direct contact with malarial parasite-infected erythrocytes are triggered for increased MIF response. One possibility is that malarial antigens that cross the placenta reach the amniotic epithelial cell interface and provide an activating stimulus. Alternatively, it could be an indirect consequence of inflammatory response elicited by PM infection. Further studies will be required to understand the biological significance of PM-induced increased MIF production on birth outcomes. Other studies have reported expression of MIF in first-trimester trophoblast and endometrium, suggesting that MIF might play a role in human implantation and in early embryonic development (2, 3). In a recent study, expression of MIF in the amnion epithelium, chorion layer, and decidua of term placenta was reported (12). The same study also reported that high concentrations of MIF were found in amniotic fluid (12). MIF levels in the amniotic fluid were higher in term placentas than in samples of amniotic fluid obtained from mid-trimester placentas, and these levels increased further during labor. Thus, a dynamic change in MIF expression levels may occur during pregnancy and labor (12). Collectively, these previous findings suggested that MIF plays an important biologic role during pregnancy, although the exact role remains to be determined.
Mononuclear cell populations in the intervillous space showed intense MIF staining. A significantly higher level of MIF expression in the mononuclear cells of PM-positive placentas suggests that these cells may be an important source for the increased MIF production observed in the IVB plasma of PM-infected placentas, as reported previously (9). Previous rodent studies have also shown that malarial parasite-infected erythrocytes can stimulate mononuclear cells to produce MIF (13). It remains to be demonstrated if increased production of MIF contributes to any antiparasitic effect mediated by mononuclear cells or other cell types.
Cytotrophoblast and Hofbauer cells expressed MIF focally. Expression of MIF in the cytotrophoblast was also reported in first-trimester placentas (2). The expression of MIF in Hofbauer cells was more frequently observed in PM-positive placentas, although it was not significantly different compared to PM-negative group. Since MIF is involved in macrophage activation, it is possible that MIF may also play a role in the activation of Hofbauer cells to protect against any organism threatening to invade the villi. At this time, it is not possible to determine what might stimulate Hofbauer cells to secrete MIF; however, if syncytiotrophoblasts were to secrete MIF into the villus stroma simultaneously with secretion into the intervillous space, then the paracrine ability of MIF to stimulate its own secretion by other surrounding cells may contribute to this pattern of Hofbauer cell expression.
Although binding of P. falciparum-infected erythrocytes to syncytiotrophoblast cells has been well documented, it was not known whether malaria parasites could modulate cytokine responses in syncytiotrophoblast cells. To further assess whether P. falciparum-infected erythrocytes can induce MIF expression, we used the choriocarcinoma cell line BeWo, which exhibits many of the known, normal functions of trophoblast cells, including hormone production. Our results clearly showed that addition of P. falciparum-infected erythrocytes to differentiated (syncytialized) BeWo cells resulted in a time-dependent, cumulative increase in MIF levels. LPS, control red blood cells, and synthetic hematin did not induce significantly higher levels of MIF than unstimulated control cultures. Coculture of BeWo cells with unselected parasites did not significantly increase MIF levels compared to normal erythrocyte stimulated control group. These findings suggest that direct binding of P. falciparum-parasitized erythrocytes and, to a lesser extent, soluble parasite factors can stimulate trophoblasts to secrete MIF. However, further studies are required to clarify the role of soluble parasite factors versus direct binding of parasites to induce production MIF. Our in vitro findings suggest that during the natural course of P. falciparum infection, the parasites and their products may be able to directly activate syncytiotrophoblast cells to secrete MIF and possibly other immune mediators. However, further studies are needed to determine if PM infection directly stimulates syncytiotrophoblasts to produce MIF in vivo.
What is the significance of having high levels of MIF in the placenta Why do different placental cellular compartments, especially those fetus-derived segments, exhibit intense MIF expression Addressing these questions will be critical for understanding the role of MIF in the biology and immunology of pregnancy, as well as its maintenance. MIF is a pluripotent factor produced by a variety of cell types. Recent studies have suggested an important role for MIF in establishing pregnancy (2, 3, 12). In this study, our focus was on the role of MIF in the context of immune responses to malaria and its potential immunomodulatory role. Given that the placenta is a rich environment for immunosuppressive hormones including glucocorticoids that can suppress phagocytosis and cytokine secretion by T lymphocytes and macrophages (4, 8), it remains a paradox how the immune system is able to clear infectious agents such as malaria parasites in the placenta. We have proposed that MIF, by virtue of its ability to overcome the immunosuppressive effect of corticoids, may play a vital role in activating macrophages to clear infectious agents like malaria parasites in the placenta (9). Although further studies will be needed to test the validity of this model for malaria and other microorganisms, it is important to recognize the potential role MIF may be playing in controlling infectious agents at the placental level. Alternatively, strong expression of MIF in the P. falciparum-infected placenta could be simply a reflection of inflammatory response that takes place in response to placental malaria infection.
In summary, we have shown that term placenta expresses MIF in various cellular compartments and that PM infection enhances its expression in mononuclear cells and amniotic epithelial cells. We have also shown that trophoblast-cytoadherent P. falciparum parasites can induce a trophoblast cell line to secrete MIF in vitro. These findings have implications for further understanding protective immune mechanisms against placental malaria.
ACKNOWLEDGMENTS
We thank all of the study participants for donating their placentae for our research efforts. We thank KEMRI and the Director of KEMRI, Davy Koech, for his approval with regard to publication of this paper. We also acknowledge the cooperation of the NNGPH Labour Ward staff and the diligent efforts of the CDC/KEMRI staff in conducting this study. We thank John Vulule, Director of the Vector Biology Control and Research Center, KEMRI, and Richard Steketee, Chief, Malaria Epidemiology Branch, DPD, CDC, for their continued support. We thank Rebecca Koopman for assistance with the parasite culture and selection work and Bryan Ihrig for his help with tissue sections. We also thank anonymous reviewers for their critical comments in revising the manuscript.
This investigation received financial support from the Medical Scholars Program of Mahidol University (to S.C.), the United Nations Development Program/World Bank/World Health Organization Special Program for Research and Training in Tropical Diseases (Principal Investigator, grant 960568 to V.U.), and the U.S. Agency for International Development (grants A0T0483-PH1-2171 and HRN-A-00-04-00010-02 to B.L.N.). J.M.M. was supported by the National Institutes of Health (NIH) (grant AI-50240) and a Faculty Grant from the University of Georgia Research Foundation. D.S.P. was supported by NIH grant AI-42318. N.L. was the recipient of University of Georgia Graduate School and Department of Infectious Diseases doctoral assistantships.
REFERENCES
1. Apte, R. S., D. Sinha, E. Mayhew, G. J. Wistow, and J. Y. Niederkorn. 1998. Cutting edge: role of macrophage migration inhibitory factor in inhibiting NK cell activity and preserving immune privilege. J. Immunol. 160:5693-5696.
2. Arcuri, F., M. Cintorino, R. Vatti, A. Carducci, S. Liberatori, and L. Paulesu. 1999. Expression of macrophage migration inhibitory factor transcript and protein by first-trimester human trophoblasts. Biol. Reprod. 60:1299-1303.
3. Arcuri, F., C. Ricci, F. Ietta, M. Cintorino, S. A. Tripodi, I. Cetin, E. Garzia, F. Schatz, P. Klemi, R. Santopietro, and L. Paulesu. 2001. Macrophage migration inhibitory factor in the human endometrium: expression and localization during the menstrual cycle and early pregnancy. Biol. Reprod. 64:1200-1205.
4. Ashwell, J. D., F. W. Lu, and M. S. Vacchio. 2000. Glucocorticoids in T cell development and function. Annu. Rev. Immunol. 18:309-345.
5. Ayisi, J. G., A. M. van Eijk, F. O. ter Kuile, M. S. Kolczak, J. A. Otieno, A. O. Misore, P. A. Kager, R. W. Steketee, and B. L. Nahlen. 2000. Risk factors for HIV infection among asymptomatic pregnant women attending an antenatal clinic in western Kenya. Int. J. STD AIDS 11:393-401.
6. Beeson, J. G., G. V. Brown, M. E. Molyneux, C. Mhango, F. Dzinjalamala, and S. J. Rogerson. 1999. Plasmodium falciparum isolates from infected pregnant women and children are associated with distinct adhesive and antigenic properties. J. Infect. Dis. 180:464-472.
7. Bernhagen, J., T. Calandra, and R. Bucala. 1998. Regulation of the immune response by macrophage migration inhibitory factor: biological and structural features. J. Mol. Med. 76:151-161.
8. Calandra, T., J. Bernhagen, C. N. Metz, L. A. Spiegel, M. Bacher, T. Donnelly, A. Cerami, and R. Bucala. 1995. MIF as a glucocorticoid-induced modulator of cytokine production. Nature 377:68-71.
9. Chaisavaneeyakorn, S., J. M. Moore, C. Othoro, J. Otieno, S. C. Chaiyaroj, Y. P. Shi, B. L. Nahlen, A. A. Lal, and V. Udhayakumar. 2002. Immunity to placental malaria. IV. Placental malaria is associated with up-regulation of macrophage migration inhibitory factor in intervillous blood. J. Infect. Dis. 186:1371-1375.
10. Egan, T. J., W. W. Mavuso, and K. K. Ncokazi. 2001. The mechanism of beta-hematin formation in acetate solution. Parallels between hemozoin formation and biomineralization processes. Biochemistry 40:204-213.
11. Fried, M., and P. E. Duffy. 1996. Adherence of Plasmodium falciparum to chondroitin sulfate A in the human placenta. Science 272:1502-1504.
12. Ietta, F., T. Todros, C. Ticconi, E. Piccoli, A. Zicari, E. Piccione, and L. Paulesu. 2002. Macrophage migration inhibitory factor in human pregnancy and labor. Am. J. Reprod. Immunol. 48:404-409.
13. Martiney, J. A., B. Sherry, C. N. Metz, M. Espinoza, A. S. Ferrer, T. Calandra, H. E. Broxmeyer, and R. Bucala. 2000. Macrophage migration inhibitory factor release by macrophages after ingestion of Plasmodium chabaudi-infected erythrocytes: possible role in the pathogenesis of malarial anemia. Infect. Immun. 68:2259-2267.
14. McGregor, I. A. 1984. Epidemiology, malaria and pregnancy. Am. J. Trop. Med. Hyg. 33:517-525.
15. Moore, J. M., B. L. Nahlen, A. Misore, A. A. Lal, and V. Udhayakumar. 1999. Immunity to placental malaria. I. Elevated production of interferon-gamma by placental blood mononuclear cells is associated with protection in an area with high transmission of malaria. J. Infect. Dis. 179:1218-1225.
16. O'Neil-Dunne, I., R. N. Achur, S. T. Agbor-Enoh, M. Valiyaveettil, R. S. Naik, C. F. Ockenhouse, A. Zhou, R. Megnekou, R. Leke, D. W. Taylor, and D. C. Gowda. 2001. Gravidity-dependent production of antibodies that inhibit binding of Plasmodium falciparum-infected erythrocytes to placental chondroitin sulfate proteoglycan during pregnancy. Infect. Immun. 69:7487-7492.
17. Ricke, C. H., T. Staalsoe, K. Koram, B. D. Akanmori, E. M. Riley, T. G. Theander, and L. Hviid. 2000. Plasma antibodies from malaria-exposed pregnant women recognize variant surface antigens on Plasmodium falciparum-infected erythrocytes in a parity-dependent manner and block parasite adhesion to chondroitin sulfate A. J. Immunol. 165:3309-3316.
18. Rocklin, R. E., J. L. Kitzmiller, and M. D. Kaye. 1979. Immunobiology of the maternal-fetal relationship. Annu. Rev. Med. 30:375-404.
19. Weinberg, E. D. 1984. Pregnancy-associated depression of cell-mediated immunity. Rev. Infect. Dis. 6:814-831.
20. Wice, B., D. Menton, H. Geuze, and A. L. Schwartz. 1990. Modulators of cyclic AMP metabolism induce syncytiotrophoblast formation in vitro. Exp. Cell Res. 186:306-316.(Sujittra Chaisavaneeyakor)
Center for Tropical and Emerging Global Diseases and Department of Infectious Diseases, College of Veterinary Medicine, University of Georgia, Athens, Georgia 30602
Department of Pathology, Egleston Children's Hospital, Emory University School of Medicine, Atlanta, Georgia 30322
Vector Biology and Control Research Center, Kenya Medical Research Institute, Kisumu, Kenya
Atlanta Research and Education Foundation, Atlanta, Georgia 30033
Department of Microbiology, Faculty of Science, Mahidol University, Bangkok 10400, Thailand
Roll Back Malaria, World Health Organization, Geneva, Switzerland
ABSTRACT
Previously, we have shown that macrophage migration inhibitory factor (MIF) was highly elevated in the placental intervillous blood (IVB) of Plasmodium falciparum-infected women. Here, we compared the expression of MIF in placental tissues obtained from P. falciparum-infected and -uninfected women. Immunoperoxidase staining showed a consistent pattern of MIF expression in syncytiotrophoblasts, extravillous trophoblasts, IVB mononuclear cells, and amniotic epithelial cells, irrespective of their malaria infection status. Cytotrophoblast, villous stroma, and Hofbauer cells showed focal staining. Only amniotic epithelial and IVB mononuclear cells from P. falciparum-infected placentas exhibited significantly higher level of MIF expression than uninfected placentas. Stimulation of syncytilized human trophoblast BeWo cells with P. falciparum-infected erythrocytes that were selected to bind these cells resulted in significant increases in MIF secretion, whereas control erythrocytes, lipopolysaccharides, and synthetic -hematin had minimal effect. These findings suggest that placental malaria modulates MIF expression in different placental compartments.
INTRODUCTION
Pregnant women are at a greater risk of acquiring Plasmodium falciparum malaria infection than nonpregnant women (14). In infected women, P. falciparum parasites accumulate in the placenta, leading to placental malaria (PM). Although both humoral and cellular immune factors have been implicated in immunity against PM (6, 11, 15-17), it remains a paradox how the immune system is able to mediate its protective effect in the presence of immunosuppressive hormones, such as glucocorticoids, which are elevated during pregnancy (18, 19).
Recently, we proposed that macrophage migration inhibitory factor (MIF), which is the only cytokine thus far known to counterregulate the immunosuppressive effect of glucocorticoids (8), may play an important role in modulating protective immune responses to infectious agents such as malaria parasites at the placental level (9). MIF is a unique cytokine in that it has properties of a cytokine, enzyme, and hormone (7). MIF was originally identified as a factor produced by T cells capable of inhibiting the random migration of mononuclear cells in vitro and is now known to be produced by several different cell types in various organs, including macrophages, lymphocytes, the pituitary, fibroblasts, and the placenta (7). MIF is released upon the activation of macrophages by various proinflammatory stimuli such as lipopolysaccharide (LPS), toxic shock syndrome toxin 1, malaria parasites, tumor necrosis factor alpha, and gamma interferon (7, 13). It is also involved in the activation of macrophages and killing of intracellular parasites (7) and in regulating natural killer cell activity (1).
Our studies demonstrated that MIF is elevated 300 to 500 fold in placental intervillous blood (IVB) plasma compared to peripheral blood plasma (9). In PM-positive women, there was a significant increase in IVB MIF levels compared to PM-negative women. We also found that placental intervillous blood mononuclear cells produced higher levels of MIF than peripheral blood mononuclear cells. Based on these observations, we have proposed that increased levels of MIF in the intervillous blood could play an important role in the activation of macrophages by overcoming the immunosuppressive effect of corticoid hormones in the placenta and thus eventually helping in the clearance of malaria parasites.
The presence of MIF protein and MIF mRNA in first-trimester human placentas was reported previously. By immunocytochemical methods, MIF was shown to be present in the placental villi, cytotrophoblasts, trophoblast cell islands, endometrium, and decidua of human first-trimester pregnancy (2). A recent study has reported the presence of MIF in the amniotic fluid, with highest levels found in term placentas (12). In the same study, MIF was also found in different cell layers of the extraembryonic membranes (12). However, a systematic investigation to characterize the expression of MIF in various cellular compartments of term placenta has been lacking.
In this study, we attempted to determine the source of MIF in various tissue compartments of term placentas and how its expression is altered in the presence of PM. In addition, we also assessed whether P. falciparum parasites can induce BeWo cells (trophoblastic origin) to secrete MIF.
MATERIALS AND METHODS
Study site, participants, and samples. Placental tissue samples were obtained from a subset of subjects who were enrolled at the New Nyanza Provincial General Hospital, Kisumu, Kenya, to participate in a cohort study to assess the impact of PM on mother-to-child human immunodeficiency virus type 1 transmission in Kenya (5). Kisumu experiences intense perennial transmission of Plasmodium falciparum malaria with two peak transmission periods occurring from November to December and May to July.
Informed consent was obtained from all patients associated with this study. The human subject experimental guidelines of the Institutional Review Board of the Centers for Disease Control and Prevention, the University of Georgia, and the Kenya Medical Research Institute Ethical Review Committee were strictly followed.
Determination of parasitemia status. The malaria infection status was confirmed by histological examination of buffered, formalin-fixed, and hematoxylin-and-eosin-stained placental sections. Placental tissues that showed the presence of malaria parasites by histological examination were categorized as PM positive, and those showed absence of parasites were categorized as the PM-negative group for this study. All the nine placentas in the PM-positive group had evidence of malaria pigment deposits. Among the nine placentas in the PM-negative groups, three had low levels of malaria pigments, indicating a past infection. However, MIF expression levels were not significantly different between the pigment-positive and pigment-negative placentas in the PM-negative group (data not shown).
Selection of placental tissues for immunohistology. Placentas collected from human immunodeficiency virus-seronegative women were selected for this study. We selected placentas from primigravidae, secundigravidae, and multigravidae in a random manner. There was no significant difference in the median age or gravidity between the PM-positive and PM-negative groups.
Collection and processing of placental tissue samples. Within 12 h of placental expulsion, several small pieces (1 by 1 by 0.3 cm) of accentric placental tissue spanning the entire thickness of the placental disk were excised and immediately placed into 10% neutral buffered formalin. Tissues were fixed overnight, then transferred to 80% ethanol, and stored until being embedded in paraffin. Placental tissue sections (3 μm thickness) obtained from sectioning the paraffin-embedded placental sample blocks were mounted onto polylysine-coated slides before MIF staining. A paraffin-embedded tonsil section from a patient with tonsillitis was used as the positive control tissue. The placental sections were stained with hematoxylin and eosin by standard procedures.
MIF immunohistochemistry. Deparaffinized and rehydrated tissue sections were placed in a moisture chamber box and incubated for 2 h in a water bath at 58°C. Antigen retrieval was performed by boiling the sections in Declere buffer (Cell Marque Corporation, Hot Springs, AR) in a stainless steel pressure cooker for 15 min. To suppress endogenous peroxidase activity, the slides were incubated for 30 min with 3% H2O2. After being washed with deionized water, samples were treated with 3% bovine serum albumin and incubated for 1 h at room temperature to block nonspecific binding. Goat anti-human MIF polyclonal antibody (R&D systems Inc, Minneapolis, MN) and purified goat immunoglobulin G isotype control antibody (Southern Biotechnology Associates, Inc., Birmingham, AL) were used at a 1:500 dilution for staining. Tissue sections were incubated with the primary antibodies overnight at 4°C and washed three times with phosphate-buffered saline (PBS). Then, the sections were incubated for 30 min at room temperature with biotinylated rabbit anti-goat antibody from the universal DAKO LSAB+ kit (DAKO Corporation, Carpinteria, CA) and washed three times with PBS. Horseradish peroxidase-labeled streptavidin from the universal DAKO LSAB+ kit was added to the slides, and the slides were incubated for 30 min at room temperature. After being washed with PBS, the slides were incubated with the chromogen substrate solution (3,3'-diaminobenzidine chromogen solution) for 5 min. After being washed with deionized water, the sections were counterstained with hematoxylin and mounted with aqueous mounting medium before being viewed under a light microscope.
Scoring system. An experienced pathologist (C.A.) examined the placental tissues, and all the sections were coded so that the reader would not know the PM infection status of samples. All sections were evaluated for the intensity of MIF staining and assigned a score ranging from 0 to 4, with the following values: 0 (negative), 1 (weak), 2 (moderate), 3 (strong), and 4 (intense strong). The distribution pattern of MIF staining was further characterized as focal (staining in some areas only), segmental (large segments were stained), or diffuse (staining in all the areas). Staining was individually evaluated in each placental compartment, i.e., amniotic epithelium, syncitio- and cytotrophoblasts, villous stromata, Hofbauer cells, and maternal decidua.
In vitro production of MIF by trophoblasts. A human choriocarcinoma cell line, BeWo, known to have characteristics of trophoblast cells, was obtained from the American Type Culture Collection (Manassas, VA). For the in vitro study, BeWo cells were first cultured in 24-well plastic culture plates (Corning) at 3 x 104 cells/well in minimal essential medium (supplemented with 10% fetal bovine serum) and treated with 40 μM of forskolin (Sigma-Aldrich, St. Louis, MO) (20) to induce syncytiotrophoblast formation. Two days later, the culture was washed once with the same culture medium and then stimulated with 3 x 106 parasitized erythrocytes; the parasites remained in the culture for the duration of the experiment. The parasite lines used for these experiments were from the 3D7 laboratory strain preselected to bind syncytialized BeWo cells by several cycles of selection (N. Lucchi et al., unpublished data). The parasites used for stimulation were semisynchronized unpurified cultures containing mainly late-stage trophozoites. These parasites were washed in culture medium twice before being used. Normal human erythrocytes (red blood cells) that were used to cultivate P. falciparum parasites were used as the control. In addition, synthetic hematin (-hematin prepared from hemin by the modified method) (100 μg/ml; Sigma-Aldrich, St. Louis, MO) (10) and 10 μg/ml of LPS (Sigma-Aldrich, St. Louis, MO) were also used to stimulate syncytialized BeWo cells. Supernatants from the cultures were collected at 0, 4, 8, 12, 24, and 48 h after stimulation and stored at –80°C until being used for measuring MIF levels.
Measurement of MIF levels in supernatants by enzyme-linked immunosorbent assay. A double sandwich enzyme-linked immunosorbent assay was used to measure MIF levels in culture supernatants as previously described (9). MIF was detected with a capture mouse monoclonal anti-human MIF antibody (clone 12302.2) and a biotinylated goat anti-human MIF antibody from R&D Systems, Inc. (Minneapolis, MN). MIF concentrations were determined using a standard curve obtained from the known concentration of cytokine standards included in each assay plate.
Statistical analyses. STATVIEW for Windows (SAS Institute, Inc., version 5) was used for data analysis, as shown in Table 1. Since data were not randomly distributed, the Mann-Whitney test was used. The Student t test was used to compare in vitro MIF data from BeWo cell experiments. P values were considered statistically significant at <0.05.
RESULTS
Immunohistochemical study of MIF. We used nine placental tissue samples from PM-negative women and nine tissue samples from PM-positive women. Immunohistochemical localization of MIF protein with a goat anti-human MIF antibody showed various degrees of positive reactivity for MIF in all placental tissues examined. The results are summarized in Table 1, and the distribution of MIF in various cellular compartments is described below.
MIF in the trophoblast layers of chorionic villi. MIF expression was commonly seen in syncytiotrophoblast layers of villi. Although its expression was widely seen in different segments of the placental sections examined, its distribution was segmental in some placentas, especially in PM-negative placentas, as shown in Fig. 1A. On the contrary, placentas from PM-positive group showed diffuse cytoplasmic staining in the syncytiotrophoblast cells (Fig. 1B). It is important to point out that syncytiotrophoblast layers from the PM-positive group (Fig. 1B) consistently exhibited intense cytoplasmic staining of MIF when compared to the PM-negative group, but the difference was not significant (Table 1).
Villous cytotrophoblast cells showed distinctly positive MIF staining. However, the staining was not uniformly expressed in all the cytotrophoblast layers of chorionic villi, and its distribution was focal. There was no appreciable difference in the pattern or intensity of MIF staining in cytotrophoblastic cells between the PM-negative (Fig. 1C) and PM-positive groups (Table 1).
MIF in the villous compartment of chorionic villi. Villous stromal cells showed MIF expression, especially in primary or stem villi (Fig. 1E). The MIF staining pattern in the villous stromal compartment did not differ between the PM-negative and PM-positive groups. Hofbauer cells (specialized fetal macrophages) in the chorionic villi of the placenta also showed positive staining with MIF (Fig. 1E and 1I). MIF staining in Hofbauer cells was generally more intense in PM-positive than in PM-negative placentas, but the difference was not statistically significant (Table 1). No MIF staining was found in the villous capillaries of chorionic villi (Fig. 1I).
MIF in intervillous blood mononuclear cells. Mononuclear cell and macrophage populations were intensely stained for MIF expression (Fig. 1D). These cell populations were commonly found in the intervillous space of PM-positive placentas. On the other hand, PM-negative placental sections had very few intervillous blood mononuclear and macrophage cell populations. Mononuclear cells in PM-positive placentas showed significantly higher intensity of MIF staining than PM-negative placentas (P < 0.02) (Table 1).
Localization of MIF at the interface of fetal and maternal compartments in placental tissue sections. Extravillous trophoblastic cells, which are fetus derived and concentrated on the maternal side of the placenta, showed an intense cytoplasmic staining of MIF. MIF expression was uniformly distributed in these cells (Fig. 1G). There was no appreciable difference in the intensity and distribution pattern of MIF staining between PM-positive and PM-negative groups.
Maternal decidual cells were generally negative for MIF expression. However, occasionally weak staining of MIF was seen focally in some decidua. Amniotic epithelial cells also uniformly showed intense staining for MIF expression (Fig. 1F). Interestingly, the amniotic epithelial cells were more intensely stained in the PM-positive group (Fig. 1F) than in the PM-negative group (Fig. 1H) (P < 0.03) (Table 1). This difference was the most notable change among various placental compartments examined in this study.
In vitro production of MIF by a choriocarcinoma cell line of trophoblastic origin Since the intensity of MIF staining was prominent in the syncytiotrophoblastic layer of PM-positive placentas compared to PM-negative placentas, we assessed the ability of syncytiotrophoblast-binding P. falciparum to induce the trophoblast cell line BeWo to produce MIF. Supernatants from unstimulated syncytialized BeWo cells showed that these cells constitutively produce MIF. MIF levels increased significantly and progressively for at least 48 h after stimulation with the P. falciparum-infected erythrocytes that were preselected to bind BeWo cells (Fig. 2). In the erythrocyte stimulated control group, secreted MIF levels did not significantly increase compared to baseline levels. Likewise, LPS and synthetic -hematin did not cause any significant increase in MIF levels compared to baseline levels.
We also determined if P.falciparum-infected erythrocytes that were preselected for binding to syncytialized BeWo cells (7XP) induced higher levels of MIF than the unselected 3D7 parental parasite line in three independent experiments. The results from these experiments showed that BeWo cells stimulated for 48 h with unselected 3D7 parasites produced significantly less MIF (arithmetic mean ± standard error, 1,629 ± 215 pg/ml) than the preselected 7XP parasites (2,671 ± 223 pg/ml) (P < 0.01). The background MIF level in the unstimulated (medium only) and normal erythrocyte-stimulated cultures were 1,065 ± 322 pg/ml and 918 ± 405 pg/ml, respectively. There was no significant difference in the MIF level between background control groups and the unselected 3D7 line-stimulated group (P > 0.05). However, the MIF level in the preselected 7XP group was significantly higher than in the background control groups (P < 0.01).
DISCUSSION
In a previous study, we demonstrated that MIF was present at very high levels in the IVB plasma compared to peripheral and cord plasma and that its level was further increased in the presence of placental malaria infection (9). Although we had shown that IVB mononuclear cells were an important source of MIF production, the contribution of other placental compartments, especially in the context of P. falciparum infection, was not known. This study has demonstrated that MIF is expressed in various cellular compartments of the term placenta, such as syncytiotrophoblasts, cytotrophoblasts, extravillous trophoblasts, Hofbauer cells, stromal cells, amniotic epithelial cells, and mononuclear cells in the intervillous space. Comparision between PM-positive and PM-negative placentas revealed significantly higher levels of MIF expression in the amniotic epithelial cells and IVB mononuclear cells. Although there was no significant quantitative difference in the syncytiotrophoblast expression of MIF between PM-infected and PM-uninfected placentas, there were some qualitative differences as seen by the uniform expression of MIF in the PM-positive placentas, while PM-negative placentas often showed segmental expression. The ability of sequestered P. falciparum parasites in the placenta to increase MIF production by syncytiotrophoblast cells was further evident from in vitro experiments, which showed that the trophoblast cell line, BeWo, specifically secreted MIF in response to cytoadherence of P. falciparum-infected erythrocytes but not other nonspecific stimuli. These observations suggest that IVB momonuclear cells and syncytiotrophoblasts could be important sources of MIF production in the placental IVB.
Amniotic epithelial cells widely showed the expression of MIF, with the PM-positive placentas showing the highest levels of MIF expression. We do not have any clear explanation of how amniotic cells that do not come in direct contact with malarial parasite-infected erythrocytes are triggered for increased MIF response. One possibility is that malarial antigens that cross the placenta reach the amniotic epithelial cell interface and provide an activating stimulus. Alternatively, it could be an indirect consequence of inflammatory response elicited by PM infection. Further studies will be required to understand the biological significance of PM-induced increased MIF production on birth outcomes. Other studies have reported expression of MIF in first-trimester trophoblast and endometrium, suggesting that MIF might play a role in human implantation and in early embryonic development (2, 3). In a recent study, expression of MIF in the amnion epithelium, chorion layer, and decidua of term placenta was reported (12). The same study also reported that high concentrations of MIF were found in amniotic fluid (12). MIF levels in the amniotic fluid were higher in term placentas than in samples of amniotic fluid obtained from mid-trimester placentas, and these levels increased further during labor. Thus, a dynamic change in MIF expression levels may occur during pregnancy and labor (12). Collectively, these previous findings suggested that MIF plays an important biologic role during pregnancy, although the exact role remains to be determined.
Mononuclear cell populations in the intervillous space showed intense MIF staining. A significantly higher level of MIF expression in the mononuclear cells of PM-positive placentas suggests that these cells may be an important source for the increased MIF production observed in the IVB plasma of PM-infected placentas, as reported previously (9). Previous rodent studies have also shown that malarial parasite-infected erythrocytes can stimulate mononuclear cells to produce MIF (13). It remains to be demonstrated if increased production of MIF contributes to any antiparasitic effect mediated by mononuclear cells or other cell types.
Cytotrophoblast and Hofbauer cells expressed MIF focally. Expression of MIF in the cytotrophoblast was also reported in first-trimester placentas (2). The expression of MIF in Hofbauer cells was more frequently observed in PM-positive placentas, although it was not significantly different compared to PM-negative group. Since MIF is involved in macrophage activation, it is possible that MIF may also play a role in the activation of Hofbauer cells to protect against any organism threatening to invade the villi. At this time, it is not possible to determine what might stimulate Hofbauer cells to secrete MIF; however, if syncytiotrophoblasts were to secrete MIF into the villus stroma simultaneously with secretion into the intervillous space, then the paracrine ability of MIF to stimulate its own secretion by other surrounding cells may contribute to this pattern of Hofbauer cell expression.
Although binding of P. falciparum-infected erythrocytes to syncytiotrophoblast cells has been well documented, it was not known whether malaria parasites could modulate cytokine responses in syncytiotrophoblast cells. To further assess whether P. falciparum-infected erythrocytes can induce MIF expression, we used the choriocarcinoma cell line BeWo, which exhibits many of the known, normal functions of trophoblast cells, including hormone production. Our results clearly showed that addition of P. falciparum-infected erythrocytes to differentiated (syncytialized) BeWo cells resulted in a time-dependent, cumulative increase in MIF levels. LPS, control red blood cells, and synthetic hematin did not induce significantly higher levels of MIF than unstimulated control cultures. Coculture of BeWo cells with unselected parasites did not significantly increase MIF levels compared to normal erythrocyte stimulated control group. These findings suggest that direct binding of P. falciparum-parasitized erythrocytes and, to a lesser extent, soluble parasite factors can stimulate trophoblasts to secrete MIF. However, further studies are required to clarify the role of soluble parasite factors versus direct binding of parasites to induce production MIF. Our in vitro findings suggest that during the natural course of P. falciparum infection, the parasites and their products may be able to directly activate syncytiotrophoblast cells to secrete MIF and possibly other immune mediators. However, further studies are needed to determine if PM infection directly stimulates syncytiotrophoblasts to produce MIF in vivo.
What is the significance of having high levels of MIF in the placenta Why do different placental cellular compartments, especially those fetus-derived segments, exhibit intense MIF expression Addressing these questions will be critical for understanding the role of MIF in the biology and immunology of pregnancy, as well as its maintenance. MIF is a pluripotent factor produced by a variety of cell types. Recent studies have suggested an important role for MIF in establishing pregnancy (2, 3, 12). In this study, our focus was on the role of MIF in the context of immune responses to malaria and its potential immunomodulatory role. Given that the placenta is a rich environment for immunosuppressive hormones including glucocorticoids that can suppress phagocytosis and cytokine secretion by T lymphocytes and macrophages (4, 8), it remains a paradox how the immune system is able to clear infectious agents such as malaria parasites in the placenta. We have proposed that MIF, by virtue of its ability to overcome the immunosuppressive effect of corticoids, may play a vital role in activating macrophages to clear infectious agents like malaria parasites in the placenta (9). Although further studies will be needed to test the validity of this model for malaria and other microorganisms, it is important to recognize the potential role MIF may be playing in controlling infectious agents at the placental level. Alternatively, strong expression of MIF in the P. falciparum-infected placenta could be simply a reflection of inflammatory response that takes place in response to placental malaria infection.
In summary, we have shown that term placenta expresses MIF in various cellular compartments and that PM infection enhances its expression in mononuclear cells and amniotic epithelial cells. We have also shown that trophoblast-cytoadherent P. falciparum parasites can induce a trophoblast cell line to secrete MIF in vitro. These findings have implications for further understanding protective immune mechanisms against placental malaria.
ACKNOWLEDGMENTS
We thank all of the study participants for donating their placentae for our research efforts. We thank KEMRI and the Director of KEMRI, Davy Koech, for his approval with regard to publication of this paper. We also acknowledge the cooperation of the NNGPH Labour Ward staff and the diligent efforts of the CDC/KEMRI staff in conducting this study. We thank John Vulule, Director of the Vector Biology Control and Research Center, KEMRI, and Richard Steketee, Chief, Malaria Epidemiology Branch, DPD, CDC, for their continued support. We thank Rebecca Koopman for assistance with the parasite culture and selection work and Bryan Ihrig for his help with tissue sections. We also thank anonymous reviewers for their critical comments in revising the manuscript.
This investigation received financial support from the Medical Scholars Program of Mahidol University (to S.C.), the United Nations Development Program/World Bank/World Health Organization Special Program for Research and Training in Tropical Diseases (Principal Investigator, grant 960568 to V.U.), and the U.S. Agency for International Development (grants A0T0483-PH1-2171 and HRN-A-00-04-00010-02 to B.L.N.). J.M.M. was supported by the National Institutes of Health (NIH) (grant AI-50240) and a Faculty Grant from the University of Georgia Research Foundation. D.S.P. was supported by NIH grant AI-42318. N.L. was the recipient of University of Georgia Graduate School and Department of Infectious Diseases doctoral assistantships.
REFERENCES
1. Apte, R. S., D. Sinha, E. Mayhew, G. J. Wistow, and J. Y. Niederkorn. 1998. Cutting edge: role of macrophage migration inhibitory factor in inhibiting NK cell activity and preserving immune privilege. J. Immunol. 160:5693-5696.
2. Arcuri, F., M. Cintorino, R. Vatti, A. Carducci, S. Liberatori, and L. Paulesu. 1999. Expression of macrophage migration inhibitory factor transcript and protein by first-trimester human trophoblasts. Biol. Reprod. 60:1299-1303.
3. Arcuri, F., C. Ricci, F. Ietta, M. Cintorino, S. A. Tripodi, I. Cetin, E. Garzia, F. Schatz, P. Klemi, R. Santopietro, and L. Paulesu. 2001. Macrophage migration inhibitory factor in the human endometrium: expression and localization during the menstrual cycle and early pregnancy. Biol. Reprod. 64:1200-1205.
4. Ashwell, J. D., F. W. Lu, and M. S. Vacchio. 2000. Glucocorticoids in T cell development and function. Annu. Rev. Immunol. 18:309-345.
5. Ayisi, J. G., A. M. van Eijk, F. O. ter Kuile, M. S. Kolczak, J. A. Otieno, A. O. Misore, P. A. Kager, R. W. Steketee, and B. L. Nahlen. 2000. Risk factors for HIV infection among asymptomatic pregnant women attending an antenatal clinic in western Kenya. Int. J. STD AIDS 11:393-401.
6. Beeson, J. G., G. V. Brown, M. E. Molyneux, C. Mhango, F. Dzinjalamala, and S. J. Rogerson. 1999. Plasmodium falciparum isolates from infected pregnant women and children are associated with distinct adhesive and antigenic properties. J. Infect. Dis. 180:464-472.
7. Bernhagen, J., T. Calandra, and R. Bucala. 1998. Regulation of the immune response by macrophage migration inhibitory factor: biological and structural features. J. Mol. Med. 76:151-161.
8. Calandra, T., J. Bernhagen, C. N. Metz, L. A. Spiegel, M. Bacher, T. Donnelly, A. Cerami, and R. Bucala. 1995. MIF as a glucocorticoid-induced modulator of cytokine production. Nature 377:68-71.
9. Chaisavaneeyakorn, S., J. M. Moore, C. Othoro, J. Otieno, S. C. Chaiyaroj, Y. P. Shi, B. L. Nahlen, A. A. Lal, and V. Udhayakumar. 2002. Immunity to placental malaria. IV. Placental malaria is associated with up-regulation of macrophage migration inhibitory factor in intervillous blood. J. Infect. Dis. 186:1371-1375.
10. Egan, T. J., W. W. Mavuso, and K. K. Ncokazi. 2001. The mechanism of beta-hematin formation in acetate solution. Parallels between hemozoin formation and biomineralization processes. Biochemistry 40:204-213.
11. Fried, M., and P. E. Duffy. 1996. Adherence of Plasmodium falciparum to chondroitin sulfate A in the human placenta. Science 272:1502-1504.
12. Ietta, F., T. Todros, C. Ticconi, E. Piccoli, A. Zicari, E. Piccione, and L. Paulesu. 2002. Macrophage migration inhibitory factor in human pregnancy and labor. Am. J. Reprod. Immunol. 48:404-409.
13. Martiney, J. A., B. Sherry, C. N. Metz, M. Espinoza, A. S. Ferrer, T. Calandra, H. E. Broxmeyer, and R. Bucala. 2000. Macrophage migration inhibitory factor release by macrophages after ingestion of Plasmodium chabaudi-infected erythrocytes: possible role in the pathogenesis of malarial anemia. Infect. Immun. 68:2259-2267.
14. McGregor, I. A. 1984. Epidemiology, malaria and pregnancy. Am. J. Trop. Med. Hyg. 33:517-525.
15. Moore, J. M., B. L. Nahlen, A. Misore, A. A. Lal, and V. Udhayakumar. 1999. Immunity to placental malaria. I. Elevated production of interferon-gamma by placental blood mononuclear cells is associated with protection in an area with high transmission of malaria. J. Infect. Dis. 179:1218-1225.
16. O'Neil-Dunne, I., R. N. Achur, S. T. Agbor-Enoh, M. Valiyaveettil, R. S. Naik, C. F. Ockenhouse, A. Zhou, R. Megnekou, R. Leke, D. W. Taylor, and D. C. Gowda. 2001. Gravidity-dependent production of antibodies that inhibit binding of Plasmodium falciparum-infected erythrocytes to placental chondroitin sulfate proteoglycan during pregnancy. Infect. Immun. 69:7487-7492.
17. Ricke, C. H., T. Staalsoe, K. Koram, B. D. Akanmori, E. M. Riley, T. G. Theander, and L. Hviid. 2000. Plasma antibodies from malaria-exposed pregnant women recognize variant surface antigens on Plasmodium falciparum-infected erythrocytes in a parity-dependent manner and block parasite adhesion to chondroitin sulfate A. J. Immunol. 165:3309-3316.
18. Rocklin, R. E., J. L. Kitzmiller, and M. D. Kaye. 1979. Immunobiology of the maternal-fetal relationship. Annu. Rev. Med. 30:375-404.
19. Weinberg, E. D. 1984. Pregnancy-associated depression of cell-mediated immunity. Rev. Infect. Dis. 6:814-831.
20. Wice, B., D. Menton, H. Geuze, and A. L. Schwartz. 1990. Modulators of cyclic AMP metabolism induce syncytiotrophoblast formation in vitro. Exp. Cell Res. 186:306-316.(Sujittra Chaisavaneeyakor)