- Email: [email protected]
Cytomegalovirus and the role of interferon in the expression of tumor necrosis factor–related apoptosis-inducing ligand in the placenta
Research
www. AJOG.org
BASIC SCIENCE: OBSTETRICS
Cytomegalovirus and the role of interferon in the expression of tumor necrosis factor–related apoptosis-inducing ligand in the placenta Janet I. Andrews, MD; Thomas S. Griffith, PhD; Jeffery L. Meier, MD OBJECTIVE: Cytomegalovirus infection causes adverse outcomes during pregnancy. Our objective was to determine the role of cytomegalovirus in modulating tumor necrosis factor (TNF)–related apoptosisinducing ligand expression in the placenta. STUDY DESIGN: TNF-related apoptosis-inducing ligand messenger RNA and protein were quantified in cytomegalovirus-infected placental fibroblasts by polymerase chain reaction and enzyme-linked immunosorbent assays, respectively. Blocking antibodies against interferon and type I interferon receptor were applied to culture medium to characterize the role of type I interferon in cytomegalovirus-induced TNFrelated apoptosis-inducing ligand upregulation. RESULTS: Expression of TNF-related apoptosis-inducing ligand mes-
senger RNA and protein was increased in cytomegalovirus-infected
placental fibroblasts, compared with uninfected controls. The cytomegalovirus-induced TNF-related apoptosis-inducing ligand messenger RNA upregulation was demonstrated across gestation, occurred in the absence of viral gene expression, and required cellular protein synthesis. TNF-related apoptosis-inducing ligand messenger RNA upregulation was markedly attenuated by inactivation of either type I interferon or its receptor. CONCLUSION: One mechanism by which cytomegalovirus infection causes unfavorable pregnancy outcomes may involve placental upregulation of TNF-related apoptosis-inducing ligand via an interferonmediated pathway.
Key words: cytomegalovirus, interferon, placenta, tumor necrosis factor-related apoptosis-inducing ligand
Cite this article as: Andrews JI, Griffith TS, Meier JL. Cytomegalovirus and the role of interferon in the expression of tumor necrosis factor–related apoptosisinducing ligand in the placenta. Am J Obstet Gynecol 2007;197:608.e1-608.e6.
H
uman cytomegalovirus (CMV) affects 1-2% of newborn infants annually in the United States, and 10-15% of infected neonates are symptomatic at birth.1 CMV infection is the leading nongenetic cause of sensorineural hearing loss in the United States.2 In addition to debilitating neurologic sequelae, CMV infection may also cause other adverse pregnancy outcomes, such as intrauterine growth restriction, birth defects, and/or fetal loss. To address this prob-
lem, the Institute of Medicine of the National Academy of Sciences has designated CMV vaccine development a top priority.3 Despite the significant morbidity associated with CMV infection, the mechanism(s) by which CMV produces these adverse outcomes is not completely understood. CMV is transmitted to the fetus via the placenta, as partly suggested by findings of CMV infection restricted to placentas of spontaneously aborted conceptuses.4
From the Departments of Obstetrics and Gynecology (Dr Andrews), Urology (Dr Griffith), and Internal Medicine (Dr Meier), Carver College of Medicine, University of Iowa, and the Veterans Affairs Medical Center (Dr Meier), Iowa City, IA. Presented at the 26th Annual Meeting of the Society for Maternal–Fetal Medicine, Miami Beach, FL, Jan. 30-Feb. 4, 2006. Received Jan. 9, 2007; revised Feb. 20, 2007; accepted April 18, 2007. Supported in part by National Institutes of Health T32 Training Grant (AI07260), the Society for Maternal–Fetal Medicine/American Association of Obstetricians and Gynecologists Foundation, and the Department of Veterans Affairs. The views expressed herein are those of the authors and do not necessarily reflect the views of the Department of Veterans Affairs or the US government. Reprints not available from the authors. 0002-9378/$32.00 © 2007 Mosby, Inc. All rights reserved. doi: 10.1016/j.ajog.2007.04.031
608.e1
American Journal of Obstetrics & Gynecology DECEMBER 2007
Examination of CMV-infected placentas from second and third trimester gestation stillbirths indicate that the majority of infected cells are fibroblasts in the stroma of the placenta.5 Most surprisingly, more than 60% of placentas obtained from healthy women undergoing either term delivery or elective pregnancy termination contained CMV proteins, underscoring that CMV is extremely successful at evading maternal immune mechanisms.6 Taken together, these findings suggest that the placenta has a major role in transmission of congenital CMV infection. The immune system plays a critical role in determining the outcome of CMV infection. CMV has developed diverse strategies to counteract host immunity. CMV-infected cells secrete many soluble factors, including mediators of inflammatory responses.7 Other strategies include alteration of expression of hundreds of cellular genes.8 An example of one such immune evasion tactic is the upregulation of cell surface
Basic Science: Obstetrics
www.AJOG.org expression of tumor necrosis factor– related apoptosis-inducing ligand (TRAIL) in CMV-infected dendritic cells resulting in the deletion or apoptosis of activated T cells.9 TRAIL is one of several tumor necrosis factor (TNF) family members capable of inducing apoptosis. We have focused on investigating CMV-induced expression of TRAIL for the following reasons: (1) induction of TRAIL expression is a proapoptotic mechanism that is initiated by other viruses, such as herpes simplex virus and human immunodeficiency virus10; (2) increased TRAIL expression has been demonstrated in CMV-infected human foreskin fibroblasts (HFF), but the mechanism behind this increase is unknown11; (3) fibroblasts from different tissues have different responses to infection,12 and it remains to be determined whether CMV modulates TRAIL expression in placenta fibroblasts; and (4) TRAIL is an interferon-stimulated gene (ISG), and many of these genes are known to be strongly induced by CMV.8 Viruses have adopted different proapoptotic strategies for immune evasion. Our understanding of how CMV might modulate proapoptotic mechanisms in the placenta to adversely affect placental immune privilege is incomplete. On the basis these findings, we hypothesized that CMV upregulates TRAIL expression during early infection to evade incoming immune cells and thereby maintain viral replication within the host cell.
M ATERIALS AND M ETHODS Cells Primary fibroblasts were isolated from first-, second-, and third-trimester placentas obtained from women undergoing either elective termination of pregnancy or uncomplicated term deliveries as approved by our institutional review board. Purified placental fibroblasts were isolated by enzymatic digestion on the basis of previous published protocols.13 After digestion, cells were cultured at 37°C in 5% CO2 by using DMEM supplemented with 10% heatinactivated fetal bovine serum and 1% streptomycin/penicillin solution. When confluent, the cells were split 1:2 ratio
and passaged 5-6 times. The homogenous cell population was characterized by positive immunostaining for vimentin. Cycloheximide, an inhibitor of TRAIL translation, was added to the culture medium at the indicated concentration. All experiments used primary placental fibroblasts unless otherwise stated. HFFs were isolated and cultured according to previously published methods.14
Reagents and antibodies The following reagents and monoclonal antibodies were used for the interferon (IFN) studies: human recombinant IFN-␣, , and ␥, and mouse monoclonal IFN-␣, , ␥, and IFN-␣/ receptor antibodies (PBL Biomedical Laboratories, Piscataway, NJ). Specific immunoglobulin G (IgG) isotype control antibodies were used when appropriate.
Viral infection For all experiments, a human Towne CMV strain that expressed green fluorescent protein (GFP) was incubated with fibroblasts at 37°C for 90 minutes. GFP expression is a marker of active infection. The virus was propagated in HFF by using methods previously described.15 Infectious CMV particles were purified by zonal centrifugation through a sorbitol density gradient to separate the virus from confounding growth factors or cytokines released into the infected cell medium, as previously described.16 The infections were performed at a multiplicity of infection (MOI) of 1-5 infectious units per cell, by using inocula with viral titers determined beforehand by conventional plaque forming assay and cytopathic effect in HFF. Inactivation of CMV by ultraviolet (UV) light was performed by previously described methods.16 UV-inactivation of viral inoculums renders the virus particles incapable of expressing the viral proteins and carrying out viral replication. Inactivation of the virus was confirmed by culturing the UV-inactivated CMV (UV-CMV) on foreskin fibroblasts and assessing viral cytopathic effect and plaque formation.
Research
RNA expression Placenta fibroblast whole-cell messenger RNA (mRNA) was isolated according to the method of Chomczynski.17 The RNA was reversed transcribed to complimentary DNA (cDNA) by using Superscript II Reverse Transcriptase (Invitrogen, Carlsbad, CA). Quantitative real-time polymerase chain reaction (Q-PCR) was performed in quadruplicate using TaqMan Gene Expression assays for TRAIL (Hs00234356) and 18S (Hs99999901; Applied Biosystems, Foster City, CA) to determine relative abundance of TRAIL mRNA after normalization to 18S mRNA. The Q-PCR was performed under standard cycling conditions for the TaqMan Gene Expression assays (ABI PRISM 7000 sequence detection system; Applied Biosystems).
Enzyme-linked immunosorbent assays After 24 hours, mock- and CMV-infected placental fibroblasts were lysed in phosphate buffered saline containing 1% Nonidet P-40 and Complete Mini, EDTA-free protease inhibitor mixture tablets (Roche Diagnostics, Mannheim, Germany). Simultaneously, cell-free fibroblast culture supernatants were collected. The lysates were centrifuged at 14,000 g to remove cellular debris before enzyme-linked immunosorbent assay (ELISA) analysis. Both fibroblast lysates and supernatants were analyzed for TRAIL production by sandwich ELISA (R&D Systems, Minneapolis, MN).
Statistical analyses All placental fibroblast experiments were performed in either duplicate or triplicate on at least 3 separate occasions. Real time reverse transcriptase-PCR data were analyzed by standard curve method after normalization to 18S. Mean (⫾ SEM) values were calculated. Data were analyzed by using Student t test and Wilcoxon signed rank test when appropriate. Differences were considered statistically significant if 2-tailed P ⬍ .05.
DECEMBER 2007 American Journal of Obstetrics & Gynecology
608.e2
Research
Basic Science: Obstetrics
R ESULTS CMV infection induces TRAIL expression in placental fibroblasts
Assessing possible mechanisms of CMV-induced TRAIL expression To elucidate the mechanism by which CMV upregulated TRAIL, we first determined whether viral or host gene expression is required for this upregulation. Because a significant quantity of fibro608.e3
FIGURE 1
TRAIL mRNA expression in mock- and CMV-infected placenta fibroblasts
A Fold increase in TRAIL mRNA
First-trimester placental fibroblasts were incubated with or without a GFP-expressing CMV strain. The entire placental fibroblast population was actively infected, as judged by the presence of characteristic CMV cytopathic effect and GFP fluorescence in all cells at 24 hours (data not shown). Cellular TRAIL mRNA expression was analyzed at 24 hours postinfection (hpi). This time point was selected for all experiments because our time course data suggested that both TRAIL mRNA and protein levels were increased at 24 hours, and that TRAIL mRNA decreased significantly between 48-72 hpi (data not shown). When compared with mock infection, TRAIL mRNA was greatly increased in the CMV-infected fibroblasts (Figure 1, A). Similarly, TRAIL protein in cellular lysate was significantly upregulated in the CMV-infected fibroblasts (Figure 1, B). The cell-free supernatants did not reveal differences in TRAIL protein expression (data not shown). To assess whether the gestational age of the placenta influenced CMV-induced TRAIL expression, fibroblasts were isolated from placentas ranging in gestational ages between 8-39 weeks. TRAIL mRNA was increased in all CMV-infected fibroblasts at 24 hpi when compared with the mock-infected group (Table 1). These results demonstrate variability in the level of TRAIL expression, which may reflect differences across both gestational age and individual placentas. We also cannot dismiss the possibility that this variation in expression could be caused by an intrinsic variability in experimental methods that cannot be precisely controlled.
www.AJOG.org
B *
*
400
6000
4000 TRAIL pg/ml
200
2000
0
0 Mock
Mock
CMV
CMV
A, Fibroblasts were harvested, cultured, and passaged from 5 first-trimester placentas. mRNA was isolated at 24 hpi and reverse transcribed. Q-PCR was performed in quadruplicate using ABI 7700 Sequence Detection System (*P ⬍ .05). B, TRAIL expression in mock- and CMV-infected placenta fibroblasts. Fibroblast cell lysates from 4 first-trimester placentas were assayed for TRAIL by ELISA (*P ⬍ .001). Andrews. Placental CMV infection and TRAIL. Am J Obstet Gynecol 2007.
blasts was required for these mechanistic studies, we first used HFF cells, because these fibroblasts were readily available, easily propagated in abundance, and, importantly, performed similarly to placental fibroblasts in exhibiting increases in TRAIL expression in response to CMV infection. HFF cells were infected with CMV or CMV inactivated by UV light (UVCMV). Notably, TRAIL mRNA (Figure 2, A) and protein (not shown) abundance was increased at 24 hpi in response to UV-CMV particles that are incapable of expressing viral gene products. This response was not observed at 4 hpi, suggesting that host cell protein synthesis was necessary. When cycloheximide was added to the CMV- and UV-CMV–infected cells, TRAIL mRNA expression was blocked (Figures 2, B and C), confirming that de novo protein synthesis was required for TRAIL upregulation. Thus, CMV activates TRAIL expression in the absence of viral gene expression, but in a manner dependent on a newly synthesized cellular protein(s). On the basis of previous reports indicating that IFN-inducible genes are among the transcripts most dramatically upregulated by CMV,18 we postulated that type I IFN (IFN-␣ and IFN-) released from CMV-infected fibroblasts
American Journal of Obstetrics & Gynecology DECEMBER 2007
may activate expression of TRAIL, an IFN-inducible gene. IFNs are classified into 2 groups, type I (IFN-␣, IFN-, and IFN-) and type II (IFN-␥), with IFN-␣ and IFN- sharing the same receptor.19 HFF cells were therefore stimulated with either human recombinant IFN-␣, IFN-, or IFN-␥ (1000 U/mL). The respective IFN blocking antibodies were added in parallel samples to determine the effectiveness of the neutralizing antibodies. The amount of TRAIL mRNA produced at 24 hours was quantified. As TABLE 1
CMV-induced TRAIL mRNA expression by gestational age Weeks gestation
Fold increase in TRAIL mRNA
8
28
8
856
10
163
10
131
11
404
18
107
38
14
39
48
........................................................................................................... ........................................................................................................... ........................................................................................................... ........................................................................................................... ........................................................................................................... ........................................................................................................... ........................................................................................................... ...........................................................................................................
TRAIL mRNA was quantified in placental fibroblasts at 24 hours postinfection. Andrews. Placental CMV infection and TRAIL. Am J Obstet Gynecol 2007.
Basic Science: Obstetrics
www.AJOG.org
Research
FIGURE 2
CMV gene expression is not required for TRAIL expression
C
B
150
100
50
10
0 Mock
CMV
Fold increase in TRAIL mRNA
Fold increase in TRAIL mRNA
Fold increase in TRAIL mRNA
A
5
0
UV-CMV
100
50
0 CMV CHX
-
+ +
+ -
UV-CMV CHX
-
+ +
+ -
A, HFF cells were mock-infected, infected with CMV or UV-CMV. mRNA was extracted at 24 hpi and analyzed for TRAIL by real-time PCR. B and C, Cycloheximide (200 g/mL) was added immediately before and during infection with CMV and UV-CMV in HFF cells. The amount of TRAIL mRNA at 24 hpi was determined in HFF cells and depicted relative to mock-infected HFF. Cellular protein synthesis was required for TRAIL upregulation in both CMV and UV-CMV infected HFF cells. Andrews. Placental CMV infection and TRAIL. Am J Obstet Gynecol 2007.
shown in Figure 3, A, TRAIL mRNA was upregulated by both IFN-␣ and IFN-, and this activity was neutralized more than 90% with the addition of the respective IFN blocking antibody. In parallel analyses, the blocking antibodies against IFN-␣ and IFN- were incubated with CMV-infected fibroblasts. QPCR analysis indicated a significant increase in TRAIL mRNA expression induced by CMV compared with the observed increase with the addition of
high units of activity of IFN alone (Figure 3, A). This IFN-mediated TRAIL expression was only partially inhibited under IFN-␣ and IFN- neutralizing conditions. While addition of IFN-␥ to HFF cells showed increased TRAIL mRNA expression, neutralizing IFN-␥ did not reduce TRAIL expression in the CMV-infected cells (not shown), indicating that IFN-␥ was unlikely to contribute to TRAIL activity in fibroblasts. This finding was expected, because fi-
FIGURE 3
Type I IFN and TRAIL mRNA expression
B CMV
300
200
100 0 0 Mock
IFNα α
IFNβ
IFNα α IFNβ +IFNα α +IFNβ Ab Ab
No Add
IFNα α IFNβ Ab Ab
% reduction in TRAIL mRNA
Fold increase in TRAIL mRNA
A
100
50
CMV upregulates TRAIL via the type I IFN pathway The role of the type I IFN signaling pathway was further explored in CMV-infected foreskin and placental fibroblasts by selectively blocking the type I IFN receptor’s response to soluble IFN-␣ and IFN-. The infected cell populations were exposed to a range of type I IFN blocking antibody concentrations (0-15 g/mL) before and throughout infection. TRAIL mRNA expression was analyzed at 24 hpi (Figure 3, B). The results indicate that CMV-induced TRAIL expression is greatly attenuated, but not completely abrogated by, type I IFN receptor blockade.
C OMMENT
0 0
5
10
15
Type 1 IFN receptor Ab (ug/mL)
A, HFF cells were cultured for 24 hours in the presence or absence of IFN-␣ (1000 U/mL) and IFN- (1000 U/mL) and antibodies to IFN-␣ (10 g/mL) and IFN- (10 g/mL). Neutralizing antibodies to IFN-␣ (10 g/mL) and IFN- (10 g/mL) were cocultured with CMV-infected HFF cells. B, Type I IFN receptor and TRAIL mRNA expression. Increasing concentrations of type I IFN receptor antibodies and nonspecific isotype control were added to placental fibroblasts prior to CMV infection. Results depict percent decrease in expression of TRAIL mRNA with varying concentrations of neutralizing antibody compared with no addition of antibody. Andrews. Placental CMV infection and TRAIL. Am J Obstet Gynecol 2007.
broblasts are not known to produce IFN-␥.20
These results indicate that CMV is a potent stimulus of TRAIL expression in placental fibroblasts during the early stage of infection. Our findings both support and extend the previous observation of Sedger et al,11 which revealed CMV-induced TRAIL expression in foreskin fibroblasts. Given that TRAIL expression is regulated in a cell-type dependent fashion,21 and that fibroblasts from different tissues can respond differently to viral infections, it was important to determine whether placental and fore-
DECEMBER 2007 American Journal of Obstetrics & Gynecology
608.e4
Research
Basic Science: Obstetrics
skin fibroblasts have similar abilities in upregulating TRAIL in response to CMV infection. Furthermore, this study differs by examining infection outcomes after inoculation with purified CMV particles to alleviate the potential for the confounding effects of the numerous soluble factors or cytokines that are known to be released from the infected cell. During early stages of infection, CMV is a potent activator of several signal transduction pathways and transcription factors, including nuclear factor-B (NF-B).22 Much evidence exists for establishing that CMV initially activates an IFN-like response mediated through the NF-B pathway.23,24 Further evidence suggests that TNF-␣ signaling pathways, through which TRAIL also operates, are inhibited during late stages of CMV-infection via inhibition of NF-B activation.25 More specifically, TRAIL-induced apoptosis can be blocked by inhibiting a NF-B binding site in the proximal TRAIL promoter.26 Taken together, this evidence leads us to hypothesize that CMV modulates TRAIL expression via type I IFN through the NF-B signaling pathway. This hypothesis is consistent with our observation that TRAIL mRNA expression was upregulated during early infection but was later downregulated at 48-72 hours. This study is limited by the absence of IFN quantification. Time course experiments to assess IFN release in CMV-infected culture medium were performed by using an ELISA method with a limit of detection of 125 pg/mL for recombinant IFN. At high multiplicity infection (MOI of 1-5), both IFN-␣ and IFN- were detected at very low levels in the supernatants across all conditions (data not shown). Possible explanations for this lack of detectable soluble IFN include: (1) IFN may be readily taken up by infected cells after its release; (2) IFN may be sequestered, masked, or degraded in the supernatant of infected cells; or (3) IFN release may be attenuated by CMV infection, yet enough IFN is made to stimulate the IFN receptors. Previous reports indicate that the amount of IFN released into CMV-infected culture supernatants may vary according to viral MOI. IFN- is generally not detected in super608.e5
natants of fibroblasts infected with high CMV titers, and IFN- is only produced at low levels by fibroblasts exposed to CMV at low MOI.27 Lee et al23 demonstrated a significant increase in IFN- protein at 12 hpi by ELISA analysis; however, these cells were infected at a low MOI. An intriguing possibility is that CMV infection might enhance IFN-dependent TRAIL upregulation through mechanisms that involve the sensitization of the infected fibroblast to low levels of IFN and the activation of other signaling or transcription regulatory pathways. This consideration is strengthened by our observations that CMV infection in conjunction with an intact IFN signaling pathway appears to be more effective than high concentrations of IFN alone in stimulating TRAIL upregulation, and that complete blocking of either type I IFN or its receptor does not fully inhibit TRAIL expression. The notion of CMV enhancing TRAIL expression through an additional mechanism accords with recently reported findings of IFN regulatory factor 3 (IRF3), a transcriptional coactivator, having IFN-independent roles in CMVinduced activation of expression of the ISG family (TRAIL is a family member) in foreskin fibroblasts28 and in paramyxovirus-induced activation of TRAIL expression in a human cell line.29 Lastly, it is conceivable that another soluble factor might also interact with IFN signaling to upregulate TRAIL expression in infected fibroblasts and, perhaps, in neighboring uninfected cells. Further studies are needed to evaluate these possibilities. Although most infected cells in a CMVinfected placenta in vivo are fibroblasts, it is not yet recognized whether TRAIL is upregulated in other placental cells in response to CMV infection. Syncytiotrophoblasts, cytotrophoblasts, and capillary endothelium cells in the placenta all express TRAIL consistently throughout gestation.30 Although we have focused this investigation on placental fibroblasts, our current studies are directed toward examining the effect of CMV on interactions between placental fibroblasts and trophoblasts. We hope to determine whether CMV directly or indirectly modulates
American Journal of Obstetrics & Gynecology DECEMBER 2007
www.AJOG.org TRAIL expression on other placental cells. Our unpublished in vitro data suggest that primary cytotrophoblasts are more resistant to CMV-induced TRAIL upregulation compared with placental fibroblasts. These differences may reflect the efficiency of the virus infection. Different CMV strains, such as laboratory-adapted vs clinical strains, may also have different influences on TRAIL expression. In conclusion, these results reinforce the evidence that the placenta plays a role in CMV transmission during pregnancy. We suggest that the virus evades the maternal immune response initially by upregulation of TRAIL via an IFN-mediated mechanism. Elucidating these mechanisms in the placenta will enable a better understanding of how CMV initiates the consequences of congenital disease and also provide the foundation for the development of prophylactic and f therapeutic intervention. ACKNOWLEDGMENTS We gratefully acknowledge Julie Mowers, Patrick McGonagill, Allen Wu, Troy Kemp, and Tamara Kucaba for their expertise and technical assistance.
REFERENCES 1. Stagno S, Pass RF, Dworsky ME, Alford CA Jr. Maternal cytomegalovirus infection and perinatal transmission. Clin Obstet Gynecol 1982; 25:563-76. 2. Fowler KB, McCollister FP, Dahle AJ, Boppana S, Britt WJ, Pass RF. Progressive and fluctuating sensorineural hearing loss in children with asymptomatic congenital cytomegalovirus infection. J Pediatr 1997;130:624-30. 3. Plotkin SA. Congenital cytomegalovirus infection and its prevention. Clin Infect Dis 2004;38:1038-9. 4. Hayes K, Gibas H. Placental cytomegalovirus infection without fetal involvement following primary infection in pregnancy. J Pediatr 1971; 79:401-5. 5. Sinzger C, Muntefering H, Loning T, Stoss H, Plachter B, Jahn G. Cell types infected in human cytomegalovirus placentitis identified by immunohistochemical double staining. Virchows Arch A Pathol Anat Histopathol 1993;423: 249-56. 6. McDonagh S, Maidji E, Chang HT, Pereira L. Patterns of human cytomegalovirus infection in term placentas: a preliminary analysis. J Clin Virol 2006;35:210-5. 7. Fortunato EA, McElroy AK, Sanchez I, Spector DH. Exploitation of cellular signaling and regulatory pathways by human cytomegalovirus. Trends Microbiol 2000;8:111-9.
www.AJOG.org 8. Zhu H, Cong JP, Mamtora G, Gingeras T, Shenk T. Cellular gene expression altered by human cytomegalovirus: global monitoring with oligonucleotide arrays. Proc Natl Acad Sci U S A 1998;95:14470-5. 9. Raftery MJ, Schwab M, Eibert SM, Samstag Y, Walczak H, Schonrich G. Targeting the function of mature dendritic cells by human cytomegalovirus: a multilayered viral defense strategy. Immunity 2001;15:997-1009. 10. Miura Y, Misawa N, Maeda N, et al. Critical contribution of tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) to apoptosis of human CD4⫹ T cells in HIV-1-infected huPBL-NOD-SCID mice. J Exp Med 2001;193: 651-60. 11. Sedger LM, Shows DM, Blanton RA, et al. IFN-gamma mediates a novel antiviral activity through dynamic modulation of TRAIL and TRAIL receptor expression. J Immunol 1999; 163:920-6. 12. Preston CM, Harman AN, Nicholl MJ. Activation of interferon response factor-3 in human cells infected with herpes simplex virus type 1 or human cytomegalovirus. J Virol 2001;5: 8909-16. 13. Fant ME. In vitro growth rate of placental fibroblasts is developmentally regulated. J Clin Invest 1991;88:1697-702. 14. Rodriguez JE, Loepfe TR, Swack NS. Beta interferon production in primed and unprimed cells infected with human cytomegalovirus. Arch Virol 1987;94:177-89. 15. Meier JL, Stinski MF. Effect of a modulator deletion on transcription of the human cytomegalovirus major immediate-early genes in infected undifferentiated and differentiated cells. J Virol 1997;71:1246-55.
Basic Science: Obstetrics 16. Meier JL, Keller MJ, McCoy JJ. Requirement of multiple cis-acting elements in the human cytomegalovirus major immediate-early distal enhancer for viral gene expression and replication. J Virol 2002;76:313-26. 17. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:156-9. 18. Boehme KW, Singh J, Perry ST, Compton T. Human cytomegalovirus elicits a coordinated cellular antiviral response via envelope glycoprotein B. J Virol 2004;78:1202-11. 19. Kayagaki N, Yamaguchi N, Nakayama M, Eto H, Okumura K, Yagita H. Type I interferons (IFNs) regulate tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) expression on human T cells: a novel mechanism for the antitumor effects of type I IFNs. J Exp Med 1999;189:1451-60. 20. Goodbourn S, Didcock L, Randall RE. Interferons: cell signalling, immune modulation, antiviral response and virus countermeasures. J Gen Virol 2000;81:2341-64. 21. Boehrer S, Nowak D, Hoelzer D, Mitrou PS, Chow KU. The molecular biology of TRAIL-mediated signaling and its potential therapeutic exploitation in hematopoietic malignancies. Curr Med Chem 2006;13: 2091-100. 22. Kowalik TF, Wing B, Haskill JS, Azizkhan JC, Baldwin AS Jr., Huang ES. Multiple mechanisms are implicated in the regulation of NFkappa B activity during human cytomegalovirus infection. Proc Natl Acad Sci U S A 1993; 90:1107-11. 23. Lee GC, Yi HA, Lee CH. Stimulation of interferon-beta gene expression by human cytomegalovirus via nuclear factor kappa B and
Research
phosphatidylinositol 3-kinase pathway. Virus Res 2006;117:209-14. 24. Yurochko AD, Kowalik TF, Huong SM, Huang ES. Human cytomegalovirus upregulates NF-kappa B activity by transactivating the NF-kappa B p105/p50 and p65 promoters. J Virol 1995;69:5391-400. 25. Jarvis MA, Borton JA, Keech AM, et al. Human cytomegalovirus attenuates interleukin-1beta and tumor necrosis factor alpha proinflammatory signaling by inhibition of NF-kappaB activation. J Virol 2006;80: 5588-98. 26. Baetu TM, Kwon H, Sharma S, Grandvaux N, Hiscott J. Disruption of NF-kappaB signaling reveals a novel role for NF-kappaB in the regulation of TNF-related apoptosis-inducing ligand expression. J Immunol 2001;167: 3164-73. 27. Abate DA, Watanabe S, Mocarski ES. Major human cytomegalovirus structural protein pp65 (ppUL83) prevents interferon response factor 3 activation in the interferon response. J Virol 2004;78:10995-1006. 28. DeFilippis VR, Robinson B, Keck TM, Hansen SG, Nelson JA, Fruh KJ. Interferon regulatory factor 3 is necessary for induction of antiviral genes during human cytomegalovirus infection. J Virol 2006;80: 1032-7. 29. Kirshner JR, Karpova AY, Kops M, Howley PM. Identification of TRAIL as an interferon regulatory factor 3 transcriptional target. J Virol 2005;79:9320-4. 30. Chen L, Liu X, Zhu Y, Cao Y, Sun L, Jin B. Localization and variation of TRAIL and its receptors in human placenta during gestation. Life Sci 2004;74:1479-86.
DECEMBER 2007 American Journal of Obstetrics & Gynecology
608.e6
www. AJOG.org
BASIC SCIENCE: OBSTETRICS
Cytomegalovirus and the role of interferon in the expression of tumor necrosis factor–related apoptosis-inducing ligand in the placenta Janet I. Andrews, MD; Thomas S. Griffith, PhD; Jeffery L. Meier, MD OBJECTIVE: Cytomegalovirus infection causes adverse outcomes during pregnancy. Our objective was to determine the role of cytomegalovirus in modulating tumor necrosis factor (TNF)–related apoptosisinducing ligand expression in the placenta. STUDY DESIGN: TNF-related apoptosis-inducing ligand messenger RNA and protein were quantified in cytomegalovirus-infected placental fibroblasts by polymerase chain reaction and enzyme-linked immunosorbent assays, respectively. Blocking antibodies against interferon and type I interferon receptor were applied to culture medium to characterize the role of type I interferon in cytomegalovirus-induced TNFrelated apoptosis-inducing ligand upregulation. RESULTS: Expression of TNF-related apoptosis-inducing ligand mes-
senger RNA and protein was increased in cytomegalovirus-infected
placental fibroblasts, compared with uninfected controls. The cytomegalovirus-induced TNF-related apoptosis-inducing ligand messenger RNA upregulation was demonstrated across gestation, occurred in the absence of viral gene expression, and required cellular protein synthesis. TNF-related apoptosis-inducing ligand messenger RNA upregulation was markedly attenuated by inactivation of either type I interferon or its receptor. CONCLUSION: One mechanism by which cytomegalovirus infection causes unfavorable pregnancy outcomes may involve placental upregulation of TNF-related apoptosis-inducing ligand via an interferonmediated pathway.
Key words: cytomegalovirus, interferon, placenta, tumor necrosis factor-related apoptosis-inducing ligand
Cite this article as: Andrews JI, Griffith TS, Meier JL. Cytomegalovirus and the role of interferon in the expression of tumor necrosis factor–related apoptosisinducing ligand in the placenta. Am J Obstet Gynecol 2007;197:608.e1-608.e6.
H
uman cytomegalovirus (CMV) affects 1-2% of newborn infants annually in the United States, and 10-15% of infected neonates are symptomatic at birth.1 CMV infection is the leading nongenetic cause of sensorineural hearing loss in the United States.2 In addition to debilitating neurologic sequelae, CMV infection may also cause other adverse pregnancy outcomes, such as intrauterine growth restriction, birth defects, and/or fetal loss. To address this prob-
lem, the Institute of Medicine of the National Academy of Sciences has designated CMV vaccine development a top priority.3 Despite the significant morbidity associated with CMV infection, the mechanism(s) by which CMV produces these adverse outcomes is not completely understood. CMV is transmitted to the fetus via the placenta, as partly suggested by findings of CMV infection restricted to placentas of spontaneously aborted conceptuses.4
From the Departments of Obstetrics and Gynecology (Dr Andrews), Urology (Dr Griffith), and Internal Medicine (Dr Meier), Carver College of Medicine, University of Iowa, and the Veterans Affairs Medical Center (Dr Meier), Iowa City, IA. Presented at the 26th Annual Meeting of the Society for Maternal–Fetal Medicine, Miami Beach, FL, Jan. 30-Feb. 4, 2006. Received Jan. 9, 2007; revised Feb. 20, 2007; accepted April 18, 2007. Supported in part by National Institutes of Health T32 Training Grant (AI07260), the Society for Maternal–Fetal Medicine/American Association of Obstetricians and Gynecologists Foundation, and the Department of Veterans Affairs. The views expressed herein are those of the authors and do not necessarily reflect the views of the Department of Veterans Affairs or the US government. Reprints not available from the authors. 0002-9378/$32.00 © 2007 Mosby, Inc. All rights reserved. doi: 10.1016/j.ajog.2007.04.031
608.e1
American Journal of Obstetrics & Gynecology DECEMBER 2007
Examination of CMV-infected placentas from second and third trimester gestation stillbirths indicate that the majority of infected cells are fibroblasts in the stroma of the placenta.5 Most surprisingly, more than 60% of placentas obtained from healthy women undergoing either term delivery or elective pregnancy termination contained CMV proteins, underscoring that CMV is extremely successful at evading maternal immune mechanisms.6 Taken together, these findings suggest that the placenta has a major role in transmission of congenital CMV infection. The immune system plays a critical role in determining the outcome of CMV infection. CMV has developed diverse strategies to counteract host immunity. CMV-infected cells secrete many soluble factors, including mediators of inflammatory responses.7 Other strategies include alteration of expression of hundreds of cellular genes.8 An example of one such immune evasion tactic is the upregulation of cell surface
Basic Science: Obstetrics
www.AJOG.org expression of tumor necrosis factor– related apoptosis-inducing ligand (TRAIL) in CMV-infected dendritic cells resulting in the deletion or apoptosis of activated T cells.9 TRAIL is one of several tumor necrosis factor (TNF) family members capable of inducing apoptosis. We have focused on investigating CMV-induced expression of TRAIL for the following reasons: (1) induction of TRAIL expression is a proapoptotic mechanism that is initiated by other viruses, such as herpes simplex virus and human immunodeficiency virus10; (2) increased TRAIL expression has been demonstrated in CMV-infected human foreskin fibroblasts (HFF), but the mechanism behind this increase is unknown11; (3) fibroblasts from different tissues have different responses to infection,12 and it remains to be determined whether CMV modulates TRAIL expression in placenta fibroblasts; and (4) TRAIL is an interferon-stimulated gene (ISG), and many of these genes are known to be strongly induced by CMV.8 Viruses have adopted different proapoptotic strategies for immune evasion. Our understanding of how CMV might modulate proapoptotic mechanisms in the placenta to adversely affect placental immune privilege is incomplete. On the basis these findings, we hypothesized that CMV upregulates TRAIL expression during early infection to evade incoming immune cells and thereby maintain viral replication within the host cell.
M ATERIALS AND M ETHODS Cells Primary fibroblasts were isolated from first-, second-, and third-trimester placentas obtained from women undergoing either elective termination of pregnancy or uncomplicated term deliveries as approved by our institutional review board. Purified placental fibroblasts were isolated by enzymatic digestion on the basis of previous published protocols.13 After digestion, cells were cultured at 37°C in 5% CO2 by using DMEM supplemented with 10% heatinactivated fetal bovine serum and 1% streptomycin/penicillin solution. When confluent, the cells were split 1:2 ratio
and passaged 5-6 times. The homogenous cell population was characterized by positive immunostaining for vimentin. Cycloheximide, an inhibitor of TRAIL translation, was added to the culture medium at the indicated concentration. All experiments used primary placental fibroblasts unless otherwise stated. HFFs were isolated and cultured according to previously published methods.14
Reagents and antibodies The following reagents and monoclonal antibodies were used for the interferon (IFN) studies: human recombinant IFN-␣, , and ␥, and mouse monoclonal IFN-␣, , ␥, and IFN-␣/ receptor antibodies (PBL Biomedical Laboratories, Piscataway, NJ). Specific immunoglobulin G (IgG) isotype control antibodies were used when appropriate.
Viral infection For all experiments, a human Towne CMV strain that expressed green fluorescent protein (GFP) was incubated with fibroblasts at 37°C for 90 minutes. GFP expression is a marker of active infection. The virus was propagated in HFF by using methods previously described.15 Infectious CMV particles were purified by zonal centrifugation through a sorbitol density gradient to separate the virus from confounding growth factors or cytokines released into the infected cell medium, as previously described.16 The infections were performed at a multiplicity of infection (MOI) of 1-5 infectious units per cell, by using inocula with viral titers determined beforehand by conventional plaque forming assay and cytopathic effect in HFF. Inactivation of CMV by ultraviolet (UV) light was performed by previously described methods.16 UV-inactivation of viral inoculums renders the virus particles incapable of expressing the viral proteins and carrying out viral replication. Inactivation of the virus was confirmed by culturing the UV-inactivated CMV (UV-CMV) on foreskin fibroblasts and assessing viral cytopathic effect and plaque formation.
Research
RNA expression Placenta fibroblast whole-cell messenger RNA (mRNA) was isolated according to the method of Chomczynski.17 The RNA was reversed transcribed to complimentary DNA (cDNA) by using Superscript II Reverse Transcriptase (Invitrogen, Carlsbad, CA). Quantitative real-time polymerase chain reaction (Q-PCR) was performed in quadruplicate using TaqMan Gene Expression assays for TRAIL (Hs00234356) and 18S (Hs99999901; Applied Biosystems, Foster City, CA) to determine relative abundance of TRAIL mRNA after normalization to 18S mRNA. The Q-PCR was performed under standard cycling conditions for the TaqMan Gene Expression assays (ABI PRISM 7000 sequence detection system; Applied Biosystems).
Enzyme-linked immunosorbent assays After 24 hours, mock- and CMV-infected placental fibroblasts were lysed in phosphate buffered saline containing 1% Nonidet P-40 and Complete Mini, EDTA-free protease inhibitor mixture tablets (Roche Diagnostics, Mannheim, Germany). Simultaneously, cell-free fibroblast culture supernatants were collected. The lysates were centrifuged at 14,000 g to remove cellular debris before enzyme-linked immunosorbent assay (ELISA) analysis. Both fibroblast lysates and supernatants were analyzed for TRAIL production by sandwich ELISA (R&D Systems, Minneapolis, MN).
Statistical analyses All placental fibroblast experiments were performed in either duplicate or triplicate on at least 3 separate occasions. Real time reverse transcriptase-PCR data were analyzed by standard curve method after normalization to 18S. Mean (⫾ SEM) values were calculated. Data were analyzed by using Student t test and Wilcoxon signed rank test when appropriate. Differences were considered statistically significant if 2-tailed P ⬍ .05.
DECEMBER 2007 American Journal of Obstetrics & Gynecology
608.e2
Research
Basic Science: Obstetrics
R ESULTS CMV infection induces TRAIL expression in placental fibroblasts
Assessing possible mechanisms of CMV-induced TRAIL expression To elucidate the mechanism by which CMV upregulated TRAIL, we first determined whether viral or host gene expression is required for this upregulation. Because a significant quantity of fibro608.e3
FIGURE 1
TRAIL mRNA expression in mock- and CMV-infected placenta fibroblasts
A Fold increase in TRAIL mRNA
First-trimester placental fibroblasts were incubated with or without a GFP-expressing CMV strain. The entire placental fibroblast population was actively infected, as judged by the presence of characteristic CMV cytopathic effect and GFP fluorescence in all cells at 24 hours (data not shown). Cellular TRAIL mRNA expression was analyzed at 24 hours postinfection (hpi). This time point was selected for all experiments because our time course data suggested that both TRAIL mRNA and protein levels were increased at 24 hours, and that TRAIL mRNA decreased significantly between 48-72 hpi (data not shown). When compared with mock infection, TRAIL mRNA was greatly increased in the CMV-infected fibroblasts (Figure 1, A). Similarly, TRAIL protein in cellular lysate was significantly upregulated in the CMV-infected fibroblasts (Figure 1, B). The cell-free supernatants did not reveal differences in TRAIL protein expression (data not shown). To assess whether the gestational age of the placenta influenced CMV-induced TRAIL expression, fibroblasts were isolated from placentas ranging in gestational ages between 8-39 weeks. TRAIL mRNA was increased in all CMV-infected fibroblasts at 24 hpi when compared with the mock-infected group (Table 1). These results demonstrate variability in the level of TRAIL expression, which may reflect differences across both gestational age and individual placentas. We also cannot dismiss the possibility that this variation in expression could be caused by an intrinsic variability in experimental methods that cannot be precisely controlled.
www.AJOG.org
B *
*
400
6000
4000 TRAIL pg/ml
200
2000
0
0 Mock
Mock
CMV
CMV
A, Fibroblasts were harvested, cultured, and passaged from 5 first-trimester placentas. mRNA was isolated at 24 hpi and reverse transcribed. Q-PCR was performed in quadruplicate using ABI 7700 Sequence Detection System (*P ⬍ .05). B, TRAIL expression in mock- and CMV-infected placenta fibroblasts. Fibroblast cell lysates from 4 first-trimester placentas were assayed for TRAIL by ELISA (*P ⬍ .001). Andrews. Placental CMV infection and TRAIL. Am J Obstet Gynecol 2007.
blasts was required for these mechanistic studies, we first used HFF cells, because these fibroblasts were readily available, easily propagated in abundance, and, importantly, performed similarly to placental fibroblasts in exhibiting increases in TRAIL expression in response to CMV infection. HFF cells were infected with CMV or CMV inactivated by UV light (UVCMV). Notably, TRAIL mRNA (Figure 2, A) and protein (not shown) abundance was increased at 24 hpi in response to UV-CMV particles that are incapable of expressing viral gene products. This response was not observed at 4 hpi, suggesting that host cell protein synthesis was necessary. When cycloheximide was added to the CMV- and UV-CMV–infected cells, TRAIL mRNA expression was blocked (Figures 2, B and C), confirming that de novo protein synthesis was required for TRAIL upregulation. Thus, CMV activates TRAIL expression in the absence of viral gene expression, but in a manner dependent on a newly synthesized cellular protein(s). On the basis of previous reports indicating that IFN-inducible genes are among the transcripts most dramatically upregulated by CMV,18 we postulated that type I IFN (IFN-␣ and IFN-) released from CMV-infected fibroblasts
American Journal of Obstetrics & Gynecology DECEMBER 2007
may activate expression of TRAIL, an IFN-inducible gene. IFNs are classified into 2 groups, type I (IFN-␣, IFN-, and IFN-) and type II (IFN-␥), with IFN-␣ and IFN- sharing the same receptor.19 HFF cells were therefore stimulated with either human recombinant IFN-␣, IFN-, or IFN-␥ (1000 U/mL). The respective IFN blocking antibodies were added in parallel samples to determine the effectiveness of the neutralizing antibodies. The amount of TRAIL mRNA produced at 24 hours was quantified. As TABLE 1
CMV-induced TRAIL mRNA expression by gestational age Weeks gestation
Fold increase in TRAIL mRNA
8
28
8
856
10
163
10
131
11
404
18
107
38
14
39
48
........................................................................................................... ........................................................................................................... ........................................................................................................... ........................................................................................................... ........................................................................................................... ........................................................................................................... ........................................................................................................... ...........................................................................................................
TRAIL mRNA was quantified in placental fibroblasts at 24 hours postinfection. Andrews. Placental CMV infection and TRAIL. Am J Obstet Gynecol 2007.
Basic Science: Obstetrics
www.AJOG.org
Research
FIGURE 2
CMV gene expression is not required for TRAIL expression
C
B
150
100
50
10
0 Mock
CMV
Fold increase in TRAIL mRNA
Fold increase in TRAIL mRNA
Fold increase in TRAIL mRNA
A
5
0
UV-CMV
100
50
0 CMV CHX
-
+ +
+ -
UV-CMV CHX
-
+ +
+ -
A, HFF cells were mock-infected, infected with CMV or UV-CMV. mRNA was extracted at 24 hpi and analyzed for TRAIL by real-time PCR. B and C, Cycloheximide (200 g/mL) was added immediately before and during infection with CMV and UV-CMV in HFF cells. The amount of TRAIL mRNA at 24 hpi was determined in HFF cells and depicted relative to mock-infected HFF. Cellular protein synthesis was required for TRAIL upregulation in both CMV and UV-CMV infected HFF cells. Andrews. Placental CMV infection and TRAIL. Am J Obstet Gynecol 2007.
shown in Figure 3, A, TRAIL mRNA was upregulated by both IFN-␣ and IFN-, and this activity was neutralized more than 90% with the addition of the respective IFN blocking antibody. In parallel analyses, the blocking antibodies against IFN-␣ and IFN- were incubated with CMV-infected fibroblasts. QPCR analysis indicated a significant increase in TRAIL mRNA expression induced by CMV compared with the observed increase with the addition of
high units of activity of IFN alone (Figure 3, A). This IFN-mediated TRAIL expression was only partially inhibited under IFN-␣ and IFN- neutralizing conditions. While addition of IFN-␥ to HFF cells showed increased TRAIL mRNA expression, neutralizing IFN-␥ did not reduce TRAIL expression in the CMV-infected cells (not shown), indicating that IFN-␥ was unlikely to contribute to TRAIL activity in fibroblasts. This finding was expected, because fi-
FIGURE 3
Type I IFN and TRAIL mRNA expression
B CMV
300
200
100 0 0 Mock
IFNα α
IFNβ
IFNα α IFNβ +IFNα α +IFNβ Ab Ab
No Add
IFNα α IFNβ Ab Ab
% reduction in TRAIL mRNA
Fold increase in TRAIL mRNA
A
100
50
CMV upregulates TRAIL via the type I IFN pathway The role of the type I IFN signaling pathway was further explored in CMV-infected foreskin and placental fibroblasts by selectively blocking the type I IFN receptor’s response to soluble IFN-␣ and IFN-. The infected cell populations were exposed to a range of type I IFN blocking antibody concentrations (0-15 g/mL) before and throughout infection. TRAIL mRNA expression was analyzed at 24 hpi (Figure 3, B). The results indicate that CMV-induced TRAIL expression is greatly attenuated, but not completely abrogated by, type I IFN receptor blockade.
C OMMENT
0 0
5
10
15
Type 1 IFN receptor Ab (ug/mL)
A, HFF cells were cultured for 24 hours in the presence or absence of IFN-␣ (1000 U/mL) and IFN- (1000 U/mL) and antibodies to IFN-␣ (10 g/mL) and IFN- (10 g/mL). Neutralizing antibodies to IFN-␣ (10 g/mL) and IFN- (10 g/mL) were cocultured with CMV-infected HFF cells. B, Type I IFN receptor and TRAIL mRNA expression. Increasing concentrations of type I IFN receptor antibodies and nonspecific isotype control were added to placental fibroblasts prior to CMV infection. Results depict percent decrease in expression of TRAIL mRNA with varying concentrations of neutralizing antibody compared with no addition of antibody. Andrews. Placental CMV infection and TRAIL. Am J Obstet Gynecol 2007.
broblasts are not known to produce IFN-␥.20
These results indicate that CMV is a potent stimulus of TRAIL expression in placental fibroblasts during the early stage of infection. Our findings both support and extend the previous observation of Sedger et al,11 which revealed CMV-induced TRAIL expression in foreskin fibroblasts. Given that TRAIL expression is regulated in a cell-type dependent fashion,21 and that fibroblasts from different tissues can respond differently to viral infections, it was important to determine whether placental and fore-
DECEMBER 2007 American Journal of Obstetrics & Gynecology
608.e4
Research
Basic Science: Obstetrics
skin fibroblasts have similar abilities in upregulating TRAIL in response to CMV infection. Furthermore, this study differs by examining infection outcomes after inoculation with purified CMV particles to alleviate the potential for the confounding effects of the numerous soluble factors or cytokines that are known to be released from the infected cell. During early stages of infection, CMV is a potent activator of several signal transduction pathways and transcription factors, including nuclear factor-B (NF-B).22 Much evidence exists for establishing that CMV initially activates an IFN-like response mediated through the NF-B pathway.23,24 Further evidence suggests that TNF-␣ signaling pathways, through which TRAIL also operates, are inhibited during late stages of CMV-infection via inhibition of NF-B activation.25 More specifically, TRAIL-induced apoptosis can be blocked by inhibiting a NF-B binding site in the proximal TRAIL promoter.26 Taken together, this evidence leads us to hypothesize that CMV modulates TRAIL expression via type I IFN through the NF-B signaling pathway. This hypothesis is consistent with our observation that TRAIL mRNA expression was upregulated during early infection but was later downregulated at 48-72 hours. This study is limited by the absence of IFN quantification. Time course experiments to assess IFN release in CMV-infected culture medium were performed by using an ELISA method with a limit of detection of 125 pg/mL for recombinant IFN. At high multiplicity infection (MOI of 1-5), both IFN-␣ and IFN- were detected at very low levels in the supernatants across all conditions (data not shown). Possible explanations for this lack of detectable soluble IFN include: (1) IFN may be readily taken up by infected cells after its release; (2) IFN may be sequestered, masked, or degraded in the supernatant of infected cells; or (3) IFN release may be attenuated by CMV infection, yet enough IFN is made to stimulate the IFN receptors. Previous reports indicate that the amount of IFN released into CMV-infected culture supernatants may vary according to viral MOI. IFN- is generally not detected in super608.e5
natants of fibroblasts infected with high CMV titers, and IFN- is only produced at low levels by fibroblasts exposed to CMV at low MOI.27 Lee et al23 demonstrated a significant increase in IFN- protein at 12 hpi by ELISA analysis; however, these cells were infected at a low MOI. An intriguing possibility is that CMV infection might enhance IFN-dependent TRAIL upregulation through mechanisms that involve the sensitization of the infected fibroblast to low levels of IFN and the activation of other signaling or transcription regulatory pathways. This consideration is strengthened by our observations that CMV infection in conjunction with an intact IFN signaling pathway appears to be more effective than high concentrations of IFN alone in stimulating TRAIL upregulation, and that complete blocking of either type I IFN or its receptor does not fully inhibit TRAIL expression. The notion of CMV enhancing TRAIL expression through an additional mechanism accords with recently reported findings of IFN regulatory factor 3 (IRF3), a transcriptional coactivator, having IFN-independent roles in CMVinduced activation of expression of the ISG family (TRAIL is a family member) in foreskin fibroblasts28 and in paramyxovirus-induced activation of TRAIL expression in a human cell line.29 Lastly, it is conceivable that another soluble factor might also interact with IFN signaling to upregulate TRAIL expression in infected fibroblasts and, perhaps, in neighboring uninfected cells. Further studies are needed to evaluate these possibilities. Although most infected cells in a CMVinfected placenta in vivo are fibroblasts, it is not yet recognized whether TRAIL is upregulated in other placental cells in response to CMV infection. Syncytiotrophoblasts, cytotrophoblasts, and capillary endothelium cells in the placenta all express TRAIL consistently throughout gestation.30 Although we have focused this investigation on placental fibroblasts, our current studies are directed toward examining the effect of CMV on interactions between placental fibroblasts and trophoblasts. We hope to determine whether CMV directly or indirectly modulates
American Journal of Obstetrics & Gynecology DECEMBER 2007
www.AJOG.org TRAIL expression on other placental cells. Our unpublished in vitro data suggest that primary cytotrophoblasts are more resistant to CMV-induced TRAIL upregulation compared with placental fibroblasts. These differences may reflect the efficiency of the virus infection. Different CMV strains, such as laboratory-adapted vs clinical strains, may also have different influences on TRAIL expression. In conclusion, these results reinforce the evidence that the placenta plays a role in CMV transmission during pregnancy. We suggest that the virus evades the maternal immune response initially by upregulation of TRAIL via an IFN-mediated mechanism. Elucidating these mechanisms in the placenta will enable a better understanding of how CMV initiates the consequences of congenital disease and also provide the foundation for the development of prophylactic and f therapeutic intervention. ACKNOWLEDGMENTS We gratefully acknowledge Julie Mowers, Patrick McGonagill, Allen Wu, Troy Kemp, and Tamara Kucaba for their expertise and technical assistance.
REFERENCES 1. Stagno S, Pass RF, Dworsky ME, Alford CA Jr. Maternal cytomegalovirus infection and perinatal transmission. Clin Obstet Gynecol 1982; 25:563-76. 2. Fowler KB, McCollister FP, Dahle AJ, Boppana S, Britt WJ, Pass RF. Progressive and fluctuating sensorineural hearing loss in children with asymptomatic congenital cytomegalovirus infection. J Pediatr 1997;130:624-30. 3. Plotkin SA. Congenital cytomegalovirus infection and its prevention. Clin Infect Dis 2004;38:1038-9. 4. Hayes K, Gibas H. Placental cytomegalovirus infection without fetal involvement following primary infection in pregnancy. J Pediatr 1971; 79:401-5. 5. Sinzger C, Muntefering H, Loning T, Stoss H, Plachter B, Jahn G. Cell types infected in human cytomegalovirus placentitis identified by immunohistochemical double staining. Virchows Arch A Pathol Anat Histopathol 1993;423: 249-56. 6. McDonagh S, Maidji E, Chang HT, Pereira L. Patterns of human cytomegalovirus infection in term placentas: a preliminary analysis. J Clin Virol 2006;35:210-5. 7. Fortunato EA, McElroy AK, Sanchez I, Spector DH. Exploitation of cellular signaling and regulatory pathways by human cytomegalovirus. Trends Microbiol 2000;8:111-9.
www.AJOG.org 8. Zhu H, Cong JP, Mamtora G, Gingeras T, Shenk T. Cellular gene expression altered by human cytomegalovirus: global monitoring with oligonucleotide arrays. Proc Natl Acad Sci U S A 1998;95:14470-5. 9. Raftery MJ, Schwab M, Eibert SM, Samstag Y, Walczak H, Schonrich G. Targeting the function of mature dendritic cells by human cytomegalovirus: a multilayered viral defense strategy. Immunity 2001;15:997-1009. 10. Miura Y, Misawa N, Maeda N, et al. Critical contribution of tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) to apoptosis of human CD4⫹ T cells in HIV-1-infected huPBL-NOD-SCID mice. J Exp Med 2001;193: 651-60. 11. Sedger LM, Shows DM, Blanton RA, et al. IFN-gamma mediates a novel antiviral activity through dynamic modulation of TRAIL and TRAIL receptor expression. J Immunol 1999; 163:920-6. 12. Preston CM, Harman AN, Nicholl MJ. Activation of interferon response factor-3 in human cells infected with herpes simplex virus type 1 or human cytomegalovirus. J Virol 2001;5: 8909-16. 13. Fant ME. In vitro growth rate of placental fibroblasts is developmentally regulated. J Clin Invest 1991;88:1697-702. 14. Rodriguez JE, Loepfe TR, Swack NS. Beta interferon production in primed and unprimed cells infected with human cytomegalovirus. Arch Virol 1987;94:177-89. 15. Meier JL, Stinski MF. Effect of a modulator deletion on transcription of the human cytomegalovirus major immediate-early genes in infected undifferentiated and differentiated cells. J Virol 1997;71:1246-55.
Basic Science: Obstetrics 16. Meier JL, Keller MJ, McCoy JJ. Requirement of multiple cis-acting elements in the human cytomegalovirus major immediate-early distal enhancer for viral gene expression and replication. J Virol 2002;76:313-26. 17. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:156-9. 18. Boehme KW, Singh J, Perry ST, Compton T. Human cytomegalovirus elicits a coordinated cellular antiviral response via envelope glycoprotein B. J Virol 2004;78:1202-11. 19. Kayagaki N, Yamaguchi N, Nakayama M, Eto H, Okumura K, Yagita H. Type I interferons (IFNs) regulate tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) expression on human T cells: a novel mechanism for the antitumor effects of type I IFNs. J Exp Med 1999;189:1451-60. 20. Goodbourn S, Didcock L, Randall RE. Interferons: cell signalling, immune modulation, antiviral response and virus countermeasures. J Gen Virol 2000;81:2341-64. 21. Boehrer S, Nowak D, Hoelzer D, Mitrou PS, Chow KU. The molecular biology of TRAIL-mediated signaling and its potential therapeutic exploitation in hematopoietic malignancies. Curr Med Chem 2006;13: 2091-100. 22. Kowalik TF, Wing B, Haskill JS, Azizkhan JC, Baldwin AS Jr., Huang ES. Multiple mechanisms are implicated in the regulation of NFkappa B activity during human cytomegalovirus infection. Proc Natl Acad Sci U S A 1993; 90:1107-11. 23. Lee GC, Yi HA, Lee CH. Stimulation of interferon-beta gene expression by human cytomegalovirus via nuclear factor kappa B and
Research
phosphatidylinositol 3-kinase pathway. Virus Res 2006;117:209-14. 24. Yurochko AD, Kowalik TF, Huong SM, Huang ES. Human cytomegalovirus upregulates NF-kappa B activity by transactivating the NF-kappa B p105/p50 and p65 promoters. J Virol 1995;69:5391-400. 25. Jarvis MA, Borton JA, Keech AM, et al. Human cytomegalovirus attenuates interleukin-1beta and tumor necrosis factor alpha proinflammatory signaling by inhibition of NF-kappaB activation. J Virol 2006;80: 5588-98. 26. Baetu TM, Kwon H, Sharma S, Grandvaux N, Hiscott J. Disruption of NF-kappaB signaling reveals a novel role for NF-kappaB in the regulation of TNF-related apoptosis-inducing ligand expression. J Immunol 2001;167: 3164-73. 27. Abate DA, Watanabe S, Mocarski ES. Major human cytomegalovirus structural protein pp65 (ppUL83) prevents interferon response factor 3 activation in the interferon response. J Virol 2004;78:10995-1006. 28. DeFilippis VR, Robinson B, Keck TM, Hansen SG, Nelson JA, Fruh KJ. Interferon regulatory factor 3 is necessary for induction of antiviral genes during human cytomegalovirus infection. J Virol 2006;80: 1032-7. 29. Kirshner JR, Karpova AY, Kops M, Howley PM. Identification of TRAIL as an interferon regulatory factor 3 transcriptional target. J Virol 2005;79:9320-4. 30. Chen L, Liu X, Zhu Y, Cao Y, Sun L, Jin B. Localization and variation of TRAIL and its receptors in human placenta during gestation. Life Sci 2004;74:1479-86.
DECEMBER 2007 American Journal of Obstetrics & Gynecology
608.e6