Previous Article | Next Article ![]()
Journal of Virology, August 2003, p. 8187-8195, Vol. 77, No. 15
0022-538X/03/$08.00+0 DOI: 10.1128/JVI.77.15.8187-8195.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Department of Comprehensive Reproductive Medicine,1 Department of Immune Regulation, Graduate School,2 Department of Cell Regulation, Medical Research Institute, Tokyo Medical and Dental University, Bunkyo, Tokyo 113-8510, Japan3
Received 19 February 2003/ Accepted 5 May 2003
|
|
|---|
|
|
|---|
When HCMV infects fibroblasts, it downregulates the surface expression of class I major histocompatibility complexes (MHCs), HLA-A and -B, to protect the infected cells from attacks by antigen-specific cytotoxic T lymphocytes (4, 44, 48). HCMV US2, US3, US6, and US11 genes are responsible for the HCMV-induced downregulation of the MHCs. US6 inhibits TAP-mediated peptide translocation to endoplasmic reticulum (ER) and subsequent peptide loading of MHC molecules (3, 16, 29). US3 prevents transport of assembled MHC-antigen complexes to the cell surface and causes an accumulation of the complexes at the ER (2, 20). Furthermore, US2 and US11 gene products mediate the rapid dislocation of MHC molecules from the ER to the cytoplasm, where they are degraded by proteasomes (21, 46). However, they show the difference in their abilities to degrade MHC molecules in dendritic cells, suggesting that HCMV has adapted itself to divergent host cell types with its multiple immunoevasive strategies (39).
Placenta as the fetomaternal interface facilitates a vast amount of trafficking of the resources required for fetal development, and HCMV is thought to take this pathway to invade the fetus. Placental villi are either floating in the intervillous space filled with maternal blood or are anchored to the basal plate facing the maternal tissue called decidua, which lines the pregnant uterus (6). A villous surface is constituted of two layers of trophoblasts: the villous cytotrophoblast in the inner layer and the villous syncytiotrophoblast in the outer layer facing the intervillous space. Another type of trophoblast residing outside the villi is called extravillous cytotrophoblasts (EVT) (10). EVT differentiate from the stem cells in cell columns of anchoring villi to proliferative and then to invasive phenotypes. They migrate into the uterine interstitium and the maternal vasculature and construct a perfusion system in early pregnancy. Trophoblasts including EVT do not express HLA-A and HLA-B molecules to protect the semiallogeneic embryo from rejection of the maternal immune system (13). However, this might make trophoblasts susceptible to natural killer (NK)-mediated cell lysis. To prevent attack by maternal CD16- CD56bright NK cells abundant in decidua (10), EVT express the nonclassical class I molecule HLA-G (28, 32, 40). HLA-G was first identified in EVT and then found in other types of cells, including monocytes, thymic epithelial cells, and tumor cells (7, 37, 45, 49). Unlike HLA-A and -B, the major function of HLA-G is to protect cells from NK lysis by activating an inhibitory receptor, KIR2DL4 (33, 38). In addition to HLA-G, EVT express HLA-C and -E, which also act as ligands for inhibitory NK cell receptors, KIR2DL and CD94/NKG2A, to downregulate the cytotoxicity of NK cells (5, 23, 24). The expression of class I MHC molecules differs among subgroups of trophoblast cells. Neither villous cytotrophoblasts nor villous syncytiotrophoblasts express any class I MHC molecules (17, 18).
In placental development, EVT invade the endometrium and uterine spiral arteries. Because inefficient invasion of EVT results in infertility, miscarriage, and preeclampsia (34), it has been a great concern whether HCMV infects EVT. Earlier studies (12, 14, 15) showed that trophoblasts are permissive for HCMV infection. Recently, it was demonstrated that HCMV efficiently infects first-trimester villous cytotrophoblast in vitro and in utero (8). These results indicate that the placental trophoblasts form a major pathway of HCMV transmission from mother to fetus. Previous studies also suggested that class I MHC expression in cytotrophoblast may be downregulated by HCMV infection. Overexpression of US3 and US6, but neither US2 nor US11, can induce downregulation of HLA-C and -G in the trophoblast-derived choriocarcinoma cell line, JEG-3 (22, 42). However, JEG-3 is nonpermissive for HCMV, and the effects of viral infection are yet unknown. Another study using indirect immunofluorescence microscopy also showed that HCMV infection reduces HLA-G expression in first-trimester trophoblasts prepared with collagenase, hyaluronidase and trypsin treatment (8), but the surface expression of class I MHC molecules has not been analyzed yet. Thus, it is still unknown whether HCMV infection downregulates class I MHC expression in fully differentiated EVT.
In this study, we report that differentiated EVT, which were prepared from first-trimester placental explants, persistently express class I MHC molecules after HCMV infection. HCMV-infected EVT expressed US gene products, US2, US3, US6, and US11, but class I MHC molecules were expressed on the cell surface at a level similar to that observed in the uninfected cells. When US2, US3, US6, or US11 was overexpressed in uninfected EVT, each US protein was localized at the ER throughout the cytoplasm but was largely dissociated from the ER after HCMV infection. In contrast, in human embryonic lung fibroblasts (HEL), these US proteins were colocalized with the ER after HCMV infection, indicating their tight association. These results suggest that the US gene products does not function in HCMV-infected EVT.
|
|
|---|
HCMV infection. HCMV (Towne) was adsorbed by HEL and EVT in DMEM containing 2% FBS for 2 h. These cells were washed five times with serum-free DMEM and then cultured in the medium containing 2% FBS for 1 to 7 days. For titration of HCMV, serial dilutions of culture medium of HCMV-infected cells were used to inoculate subconfluent HEL cultures in eight-well culture slides (Becton Dickinson). These cells were cultured for 1 week and then examined for expression of UL112-113 using M23 monoclonal antibody (MAb) (19).
Indirect immunofluorescence microscopy. Cells cultured in eight-well slides were fixed with 4% paraformaldehyde in phosphate-buffered saline for 20 min at room temperature. Cells were reacted with primary antibodies in phosphate-buffered saline containing 3% bovine serum albumin for 1 h, and then with fluorescein isothiocyanate-, rhodamine- or Cy3-conjugated secondary antibodies for 1 h. Cells were examined using an Axiophot 2 (Zeiss) fluorescence microscope equipped with a Fujix HC-300 (Fuji) digital camera and an LSM510 (Zeiss) confocal laser scanning microscope.
Flow cytometry. After trypsinization, cells were incubated with anti-class I MHC MAb (1:100) or an isotype control for 30 min at 4°C and with a fluorescein isothiocyanate-conjugated F(ab')2 fraction of goat anti-mouse IgG (1:200) for 30 min at 4°C and then were analyzed with the flow cytometer FACScalibur (Becton Dickinson).
Antibodies.
Antibodies used in this study were anti-IE1/2 MAb 810 (Chemicon), anti-UL112-113 MAb M23 (9), anti-pp65 MAb 1-I-11 (ViroGen), anti-class I MHC MAb W6/32 (Dako), anti-FLAG MAb M2 (Stratagene), anticalnexin rabbit polyclonal antibody SPA-860 (StressGen), anticalreticulin goat polyclonal antibody C-17 (Santa Cruz), anticytokeratin MAb MNF116 (Dako), antivimentin MAb V9 (Nichirei), anti-integrin
1 MAb 1973 (Chemicon), anti-integrin ß1 MAb 1951 (Chemicon), anti-integrin
6 MAb CBL 458 (Cymbus), and anti-integrin ß4 MAb 2058 (Chemicon).
Transient expression of US proteins. To construct expression plasmids for the US2, US3, US6 and US11 proteins tagged with FLAG epitope at the C-terminal end, the coding regions were amplified by PCR and cloned into pBluescript II KS(+) (Stratagene) containing a FLAG sequence. The US-FLAG cDNAs were then cloned into an expression plasmid, pME18S. These expression plasmids (0.2 µg) were introduced into the cells plated on eight-well culture slides with FuGene 6 (Roche). The efficiency of the plasmid transfection to EVT and HEL in this experiment was about 1%.
RT-PCR. Total RNA (0.2 µg) of cultured cells was examined with the one-step reverse transcription (RT)-PCR kit (Invitrogen) according to the manufacturer's protocol. The primer sets used for RT-PCR were as follows: 5'-TCGTTAAAGTGGAACGTG-3' (US2 sense, the nucleotide coordinates in HCMV AD169: 193416 to 193433), 5'-ACTATTGTCCAGGCCACA-3' (US2 antisense, 193598 to 193615), 5'-CTTACATGGACAGACTGC-3' (US3 sense, 194353 to 194370), 5'-GCTGAAGGTACCAGTTGA-3' (US3 antisense, 194554 to 194571), 5'-GCACAGACCCGTTTGTTA-3' (US6 sense, 195534 to 195551), 5'-TAGCCGACGGACTCGTTG-3' (US6 antisense, 195703 to 195720), 5'-CCTGCCACCAATGCCAAA-3' (US11 sense, 200104 to 200121), and 5'-AAAATGTCGGTGCAGCCA-3' (US11 antisense, 200316 to 200333). The primer set for ß-actin mRNA was purchased from TAKARA. The RT reaction was performed at 50°C for 30 min. PCR was repeated for 30 cycles of 15 s at 94°C, 30 s at 55°C, and 1 min at 72°C with a thermal cycler GeneAmp PCR System 9700 (Applied Biosystems).
|
|
|---|
1 and ß1 (Fig. 2C) and the ability to invade through the reconstituted basement membrane, Matrigel (data not shown). We also confirmed that they expressed neither a marker of fibroblasts, vimentin, nor those of proliferative cytotrophoblasts, such as integrins
6 and ß4 (Fig. 2C).
![]() View larger version (42K): [in a new window] |
FIG. 1. Schematic view of an anchoring chorionic villus in early pregnancy. The EVT express HLA-C, -E, and -G along the differentiation process from the proliferative phenotype in the cell column (CC) to the invasive phenotype in the basal plate (BP). Abbreviations: CP, chorionic plate; DC, decidual cell; FA, fetal artery; FB, fibroblast; FV, fetal vein; HC, Hofbauer cell (placental macrophage); IVS, intervillous space; LGL: large granular lymphocyte (NK cell); UPA, uteroplacental artery; UPV, uteroplacental vein; vCTB, villous cytotrophoblast; vSTB, villous syncytiotrophoblast.
|
![]() View larger version (103K): [in a new window] |
FIG. 2. HCMV infection of the differentiated EVT. (A) Explant of a first-trimester chorionic villus; (B) isolated EVT; (C) immunocytochemical analysis of EVT. Isolated EVT were stained with specific antibodies indicated in the figure and detected by peroxidase-conjugated secondary antibodies and DAB. Scale bars: 10 µm (A) and 5 µm (B and C).
|
![]() View larger version (34K): [in a new window] |
FIG. 3. Viral gene expression and production of infectious virions in HCMV-infected EVT. (A) EVT infected with HCMV at an MOI of 5. Scale bar; 5 µm. (B) Expression of HCMV genes in infected EVT. EVT were infected with HCMV at an MOI of 5, and the expression of immediate-early (IE1/IE2), early (UL112-113), and early-late (pp65) proteins was examined by indirect immunofluorescence microscopy (C) Expression of IE1/IE2, UL112-113, and pp65 in HCMV-infected EVT and HEL. EVT and HEL were infected with HCMV at an MOI of 1 or 5 and were examined by the indirect immunofluorescence method at 24, 48, and 72 hpi. (D) Production of infectious virions in EVT and HEL. The single-step growth curve of HCMV in EVT and HEL after infection of HCMV at an MOI of 5.
|
HCMV infection did not downregulate surface expression of class I MHC molecules in EVT. EVT express the class I MHC molecules HLA-C, -E, and -G but not HLA-A and -B (13). It was previously shown by an indirect immunofluorescence method that HCMV infection suppresses expression of HLA-G in cultured first-trimester villous cytotrophoblasts (8). We first examined the expression of class I MHC molecules in HCMV-infected EVT by a similar method. EVT were infected at an MOI of 5, and their expression of the class I MHC molecules was examined using the MAb W6/32 that recognizes HLA-A, -B, -C, -E, and -G. Use of a relatively high MOI ensured that most cells (>95%) were infected with HCMV (Fig. 3C). As shown in Fig. 4A, HCMV-infected EVT expressed the class I MHC molecules at a level similar to that observed in uninfected cells. Then, we examined the surface expression of the class I MHC molecules (HLA-C, -E, and -G) by flow cytometry and found that HCMV infection did not affect their expression at 72 hpi (Fig. 4B). As a control experiment, we examined HCMV-infected HEL and confirmed that HCMV infection completely suppressed the surface expression of the class I MHC molecules (HLA-A and -B) at 72 hpi as reported previously (4, 44, 48). We also examined class I MHC expression at later stages of infection, and found that the expression was maintained 7 days after HCMV infection (Fig. 4C).
![]() View larger version (38K): [in a new window] |
FIG. 4. Expression of class I MHC molecules in HCMV-infected EVT. (A) Indirect immunofluorescence analysis. Expression of class I MHC molecules in HCMV-infected EVT. EVT were infected with HCMV at an MOI of 5 and examined at 72 hpi using the MAb W6/32, which reacts with HLA-A, -B, -C, -E, and -G. (B) Flow cytometric analysis of the surface expression of class I MHC molecules at 72 hpi. EVT express HLA-C, -E, and -G. HEL express HLA-A and -B. These class I MHC molecules of HCMV-infected cells (thick line) and mock-infected cells (thin line) were detected using the MAb W6/32 or isotype control (dotted line). (C) Expression of class I MHC molecules in HCMV-infected EVT on later days. The flow cytometry of EVT was performed for 7 days after infection. Shaded curve, W6/32 antibody; open curve, isotype control antibody.
|
![]() View larger version (52K): [in a new window] |
FIG. 5. Expression of US2, US3, US6, and US11 mRNA in HCMV-infected EVT and HEL. Total RNA (0.2 µg) was prepared from mock- or HCMV-infected (MOI = 5) cells and examined by RT-PCR.
|
![]() View larger version (60K): [in a new window] |
FIG. 6. Subcellular localization of ER in HCMV-infected EVT. (A) The localization of ER in EVT and HEL was examined using anticalnexin antibody and anticalreticulin antibody. DNA was stained with Hoechst 33258. Cells were infected at an MOI of 5 and examined at 24 hpi. (B) Localization of FLAG-tagged US proteins in EVT was examined. EVT were transfected with the expression plasmids for FLAG-tagged US proteins, US2F, US3F, US6F, and US11F. The localization of FLAG-tagged proteins was detected at 24 h after transfection using anti-FLAG antibody. The localization of the ER in the same cells was examined with anticalnexin antibody. (C) Localization of FLAG-tagged US proteins and ER (calnexin) in HCMV-infected EVT was examined. EVT were transfected with the expression plasmids for FLAG-tagged US proteins and then infected with HCMV at an MOI of 5 at 24 h after transfection. The localization of FLAG-tagged proteins was detected at 24 hpi using anti-FLAG antibody. The localization of the ER in the same cells was examined with anticalnexin antibody. (D) Localization of FLAG-tagged US proteins and ER (calreticulin) in HCMV-infected EVT. EVT were transfected with the expression plasmids and then infected with HCMV as described in Fig. 6C. The localization of FLAG-tagged proteins and the ER were detected at 24 hpi using anti-FLAG antibody and anticalreticulin antibody. All cells were examined using a confocal laser scanning microscope. DIC, differential interference contrast microscopy. Scale bar, 10 mm.
|
US2, US3, US6, and US11 proteins were colocalized with ER in HCMV-infected HEL. As a control experiment, we expressed US2F, US3F, US6F, and US11F and examined their localization in HEL, in which class I MHC is known to be downregulated by HCMV infection. In the uninfected HEL, US2F, US3F, and US11F were colocalized with the ER in the perinuclear domain (Fig. 7A). Overexpression of these US proteins resulted in the aggregation of ER at the perinuclear region because it was not observed in HCMV-infected HEL (Fig. 6A). US6F also colocalized with the ER, but it did not induce aggregation. We then examined the localization of the FLAG-tagged US proteins in the plasmid-transfected HEL after subsequent HCMV infection (Fig. 7B). In contrast to EVT, these US proteins were colocalized with the ER in most of the HCMV-infected HEL (77 to 86%), indicating that they were tightly associated with or integrated into the ER membrane.
![]() View larger version (49K): [in a new window] |
FIG. 7. Subcellular localization of ER in HCMV-infected HEL. (A) Localization of FLAG-tagged US proteins in HEL was examined. HEL were transfected with the expression plasmids for FLAG-tagged US proteins, i.e., US2F, US3F, US6F, and US11F, and FLAG-tagged proteins were detected at 24 h after transfection. (B) Localization of FLAG-tagged US proteins in HCMV-infected HEL was examined at 24 hpi. HEL were transfected with the expression plasmids for FLAG-tagged US proteins as described for panel A and then infected with HCMV at an MOI of 5 at 24 h after transfection. The localization of FLAG-tagged proteins and the ER was detected using anti-FLAG antibody and anticalnexin antibody. These cells were examined using a confocal laser scanning microscope. DIC; differential interference contrast microscopy. Scale bar, 10 mm.
|
|
|
|---|
We showed that EVT continued to express class I MHC molecules after HCMV infection. However, this result does not imply that the HLA-C, -E, and -G molecules themselves are resistant to the HCMV-mediated downregulation, because the surface expression of HLA-C and -G was suppressed by US3 and US6 in JEG-3 cells (22). The difference between HEL and EVT in expression levels of US2, US3, US6, and US11 mRNA was too small to account for the resistance of EVT to the downregulation of class I MHC (Fig. 5). It is thus likely that these US proteins are expressed but do not function in EVT. US6 inhibits the TAP-mediated peptide translocation to ER (3, 16, 29), and US3 prevents the transport of assembled MHC-antigen complexes (2, 20). US2 and US11 mediate a rapid dislocation of MHC molecules from the ER to the cytoplasm (21, 46). All these US products are type I transmembrane glycoproteins, their integration into the ER membrane being essential for their functions. In this study, we found that these US gene products were largely displaced from ER in HCMV-infected EVT (Fig. 6C and D). This indicates that the products do not effectively associate with or integrate into the ER, thus allowing the surface expression of class I MHC molecules.
In a previous study suggesting downregulation of HLA-G in HCMV-infected cytotrophoblasts, cells were prepared with enzymatic digestion of chorionic villi (8). We examined differentiated EVT that had grown out of villous explants by immunofluorescence microscopy and flow cytometer analysis and found that HCMV infection did not reduce expression of class I MHC molecules (Fig. 4). It has been widely accepted that EVT differentiate through several stages after leaving the cell columns (10). The EVT located on and near the basal lamina show a proliferative phenotype, and those left there gradually become invasive. It is likely that the cytotrophoblasts prepared by enzymatic digestion include a wide spectrum of cells at various stages, to which the difference of MHC class I sensitivity may be attributed. We noticed that the morphology was also significantly different among these trophoblast cells. The cytotrophoblasts and JEG-3 cells sensitive to the MHC downregulation show a typical polygonal morphology of epithelial cells with a small cytoplasm and a large nucleus. In contrast, the differentiated EVT used in our study have a large and extended cytoplasm relative to the nucleus (Fig. 2 and 6A). A subset of cytotrophoblasts, presumably with the proliferative phenotype, may be sensitive to the HCMV-mediated MHC downregulation, but the differentiated EVT are not. Although we demonstrated that the total amount of the class I MHC molecules in EVT was not affected by HCMV infection, we did not use specific antibodies against either HLA-C, -E, or -G in this study. Therefore, it is possible that the ratio of HLA-C, -E, or -G molecules expressed in EVT may change after HCMV infection.
The ER is known to associate directly with microtubules but not with actin filaments (25, 30, 43). Disruption of microtubules by nocodazole induces retraction of the ER network toward the cell center, indicating that the ER is delivered by microtubule-dependent machinery. Overexpression of the US2, US3, or US11 gene in HEL induced an aggregation of the ER (Fig. 7A) similar to that caused by microtubule disruption (43), suggesting that an excess of US proteins affects the interaction between the ER and microtubules. Interestingly, HCMV infection of EVT induced ER relocation toward the cell membrane in the opposite direction to that induced by the US proteins in HEL (Fig. 6A). Microtubules are oriented with their minus ends anchored in the centrosome and their plus ends toward the cell membrane, and they transport membrane vesicles and organelles in either direction using motor proteins such as kinesin and dynein. Therefore, EVT ER may associate with microtubules via molecular machinery different from that used in fibroblasts.
The first-trimester EVT infected with HCMV face maternal NK cells, which dominate the leukocytes of decidua by 70% (10). If HCMV downregulates HLA-C, -E, and -G on EVT, the infected cells should be targets of decidual NK cells, and this may trigger a full-scale response of the maternal immune system against the placenta. Therefore, the persistent expression of these MHC molecules in HCMV-infected EVT may help the virus to survive in the fetus and the placenta. Expression of HLA-G was first detected in EVT (28) and then was also discovered in monocytes where HCMV established a latent and persistent infection (49). Interestingly, HLA-G is upregulated during HCMV reactivation in macrophages differentiated from latently infected monocytes (35, 36). These results suggest that HCMV may have adapted itself to survive in HLA-G-expressing cells, taking advantage of HLA-G to evade the host immune system.
This work is supported by a grant-in-aid for scientific research of the Japan Society for the Promotion of Science (JSPS).
|
|
|---|
. J. Immunol. 156:4224-4231.[Abstract]
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»