Human Cytomegalovirus Enhances Placental Susceptibility and Replication of Human Immunodeficiency Virus Type 1 (HIV-1), Which May Facilitate In Utero HIV-1 Transmission

Co-pathogens that threaten the fetus during gestation, including human cytomegalovirus (HCMV), may contribute to mother-to-child transmission of HIV-1. We show that HCMV infection of placental macrophages may promote HIV replication and transmission as a consequence of inflammation and inhibition of intrinsic antiviral responses.

Abstract

Several co-pathogens that pose threats to the fetus during gestation, including human cytomegalovirus (HCMV), may also contribute to mother-to-child transmission (MTCT) of human immunodeficiency virus type 1 (HIV-1). Within endemic settings, associations between maternal HCMV viral load and increased incidence of MTCT of HIV-1 are documented; however, the mechanisms that promote transmission are poorly characterized. Here we demonstrate that HCMV coinfection enhances susceptibility and viral replication of HIV-1 in placental macrophages (Hofbauer cells) in vitro. Consistent with enhanced viral susceptibility, HCMV exposure upregulates CCR5 and CD80 expression on Hofbauer cells. HCMV also significantly induces type I interferon (IFN), proinflammatory cytokines, and antiviral gene expression. Interestingly, we found that HCMV diminishes type I IFN–mediated phosphorylation of STAT2. Collectively, our data suggest that HCMV-induced activation, local inflammation, and antagonism of type I IFN responses in placental Hofbauer cells promote in utero transmission of HIV-1.

Even with optimal adherence, maternal antiretroviral therapy reduces, but does not eliminate, vertical transmission of human immunodeficiency virus type 1 (HIV-1) following in utero, intrapartum, and postpartum exposure [1, 2]. A potential barrier for elimination, particularly in resource-poor settings, is maternal coinfection during pregnancy, which may facilitate in utero transmission of HIV-1 [3–5]. Along with HIV-1, human cytomegalovirus (HCMV) is the most common viral agent transmitted from mother to child [6]. More than 90% of reproductive-aged women in developing countries, and <50% in many Western settings, are currently infected with HCMV [7]. In addition, HCMV coinfection and reactivation are common in pregnant women living with HIV. Interestingly, there is emerging clinical (in utero and intrapartum) and in vitro data associating HCMV with mother-to-child transmission (MTCT) of HIV-1 [5, 8–13]. In particular, a recent study examined the timing of HCMV infection relative to HIV-1 infection and found that in utero HCMV infection correlated with both in utero and intrapartum HIV-1 infection [5], suggesting that fetal HCMV infection may predispose infants to in utero HIV-1 infection.

Studies from our group have demonstrated that stimulation of fetal mononuclear cells with HCMV increased expression of CCR5, suggesting a mechanism by which HCMV might increase fetal susceptibility to HIV-1 [11]. In addition, local inflammation leading to trophoblast damage [14] may also provide HIV-1 access in placental target cells and enhance viral replication. HCMV directly upregulates expression of inflammatory mediators to maintain an environment that is beneficial for its own replication, and also leads to rapid induction of the type I interferon (IFN) pathway, concurrently contributing to an antiviral state in infected and surrounding tissues [15, 16]. This dichotomy is underscored by data showing that, although HCMV induces a rapid type I IFN response, it generates inhibitory antiviral and immunoregulatory effects at multiple steps [17, 18] by degrading JAK1 [18], interfering with STAT signaling [19], and preventing antiviral IFN-stimulated gene production [20].

HIV-1 and HCMV must cross the placenta to infect the fetus in utero. The placenta consists of numerous macrophages (Hofbauer cells), which are thought to be key mediators involved in HIV-1 and HCMV vertical transmission to the fetus [21–23]. Ex vivo studies have documented that during pregnancy HCMV infects the placenta before the fetus [21, 23, 24] as a consequence of maternal viremia [22, 25]. We postulate that HCMV may promote placental HIV-1 replication and in utero HIV-1 transmission as a consequence of inflammation and inhibition of intrinsic antiviral responses. Here we show that HCMV infection of placental Hofbauer cells upregulates CCR5 and CD80 expression, induces cellular activation and expression of tumor necrosis factor alpha (TNF-α), interleukin 6 (IL-6), and type I IFNs, and inhibits STAT2 expression and phosphorylation, which may contribute to observed increased HIV-1 susceptibility and replication in Hofbauer cells. These observations emphasize the significance that HCMV-induced activation may have on target cells at the placenta resulting in fetal HIV-1 infection. These studies also provide a foundation upon which to design strategies for optimal antenatal prophylaxis in settings where maternal HIV-1/HCMV coinfection is endemic.

METHODS

Ethics Statement

With written informed consent, term (>37 weeks’ gestation) placentae from healthy women (>18 years of age) were obtained immediately following cesarean delivery from Grady Memorial and Emory Midtown Hospitals in Atlanta, Georgia. Approval of the study was granted from the Emory University Institutional Review Board and the Grady Research Oversight Committee. Samples were de-identified prior to handling by laboratory personnel.

Isolation of Hofbauer Cells

To isolate Hofbauer cells, villous tissue was dissected from the placenta, as previously described [21, 26]. The tissue was mechanically dispersed and resuspended in Hank’s balanced salt solution containing 10% Trypsin/ethylenediaminetetraacetic acid (EDTA) (Sigma-Aldrich) and 1 mg/mL of DNAse I (Sigma-Aldrich) for 45 minutes, followed by resuspension in media containing 1 mg/mL collagenase IV (Worthington Biochemical) and 0.2 mg/mL of DNAse I (Sigma-Aldrich), and incubated in a shaking water bath at 37°C for 1 hour. The digested tissue was passed through gauze and a 40-µm cell strainer (BD-Falcon Biosciences). The mononuclear cell population was isolated by density gradient centrifugation on Histopaque-1077 (Sigma-Aldrich). CD14⁺ magnetic cell sorting was performed using anti-CD14 magnetic beads (Miltenyi Biotech) as recommended by the manufacturer. The purity of the Hofbauer cell population was assessed by CD14 staining and was on average >97%.

Virus Infection of Hofbauer Cells

HIV-1 and HCMV infections of Hofbauer cells were performed as previously described [21, 24, 27]. For HIV-1 infections, 1.0 × 10⁵ cells/well in a 96-well plate (Corning) were infected at a multiplicity of infection (MOI) of 0.1 overnight at 37°C with the HIV-1 BaL strain (HIV-1_BaL). Cells were then washed with phosphate-buffered saline (PBS) to remove unabsorbed virus and replenished with complete media. To monitor HIV-1 production, cell supernatants were collected at various days postinfection. Viral replication was detected by p24 released into the supernatant by enzyme-linked immunosorbent assay (ELISA) (Advanced BioScience Laboratories). The HIV-1_BaL strain is R5-trophic and was isolated from infant lung tissue [28]. For HCMV infections, the strain TB40/E-GFP was provided by Christian Sinzger (Ulm University, Germany) [29] and Don Diamond (City of Hope, Duarte, California) [27]. Then, 1.0 × 10⁵ cells/well in a 96-well plate or 5.0 × 10⁵ cells/well in a 24-well plate were infected at an MOI of 0.1 overnight at 37°C. Cells were then washed with PBS to remove unabsorbed virus and replenished with complete media. For coinfections, cells were infected with HCMV for 2 days prior to HIV-1 infection or were infected with HIV-1 for 2 days prior to HCMV infection.

Real-time Polymerase Chain Reaction

Messenger RNA (mRNA) was extracted using the RNAeasy kit (Qiagen). The complementary DNA was transcribed using QuantiTect RT kit (Qiagen). Host gene expression was performed using SYBR green with appropriate primer sets (Supplementary Table 1). All reactions were run in triplicate using the Applied Biosystems Prism 7500 Sequence Detection System. Delta cycle threshold (Ct) values from the calibrator and experimental groups were measured by subtracting Ct values from target vs the housekeeping transcript, β-actin. Gene expression data are represented as fold change relative to time-matched, mock-infected controls (gene expression normalized to β-actin − ΔΔCt method).

Flow Cytometry

Cells were washed with PBS and gently detached using 0.5 mM EDTA in PBS. Then, 5 × 10⁵ cells were labeled with the following surface-marker antibodies: CD14, CCR5, CXCR4, CD16, CD80, CD86, PDL-1, and HLA-DR (BD Biosciences). For the infection assay, green fluorescent protein (GFP)–positive cells (infected) and GFP-negative cells (uninfected) were analyzed 48 hours postinfection on a BD LSR II flow cytometer driven by the DiVA software package (Becton Dickinson). Analysis of the acquired data was performed using FlowJo software (Tree Star).

ELISA

Cytokine analysis was performed on supernatants from mock or HCMV-infected cells using TNF-α, IL-6, interleukin 10 (IL-10) (R&D Systems), IFN-α, and IFN-β (PBL) ELISAs according to the manufacturers’ instructions.

Immunoblotting

Cells were lysed in radio-immunoprecipitation assay buffer with protease inhibitors (Roche). Samples were subjected to denaturing sodium dodecyl sulfate–polyacrylamide gel electrophoresis. Gels were blotted on nitrocellulose membranes (GE Healthcare). After blocking with buffer (LI-COR), the membrane was incubated with primary antibodies (anti-STAT1, anti-pSTAT1, anti-STAT2, anti-pSTAT2 [Abcam], and anti- IE1 [provided by Don Diamond, City of Hope]) overnight at 4°C. Following incubation with appropriate secondary antibody, the immunoreactive bands were visualized and the signal intensity quantified using Odyssey Imaging system (LI-COR). The blots were reprobed with anti–β-actin to correct for differences in protein loading.

Statistical Analysis

All figures are representative of at least 3 independent experiments and 10 individual donors. Data were analyzed by 1-way analysis of variance (ANOVA) followed by Dunnett test for multiple comparisons or by 2-way ANOVA followed by Sidak test for multiple comparisons. Differences were defined as significant when P ≤ .05. All statistical analysis was performed using GraphPad Prism software.

RESULTS

HCMV Coinfection Increases HIV-1 Susceptibility and Replication in Placental Hofbauer Cells

During maternal HIV-1/HCMV coinfection, HCMV may serve as an independent risk factor for vertical transmission of HIV-1 [8–11]; however, the underlying mechanisms that increase MTCT remain unknown. Here we show that Hofbauer cells are permissive to HCMV (TB40/E-GFP) (Figure 1A). To determine whether HCMV infection is associated with greater susceptibility to HIV-1 infection and viral replication in placental target cells, we compared viral replication in Hofbauer cells infected with HCMV (TB40/E-GFP) 48 hours before and after infection with HIV-1_BaL. p24 levels were detected in supernatant over time. Mean production of p24 antigen in both groups of Hofbauer cells coinfected with HCMV and HIV-1 was significantly elevated, compared with production in control cells cultured in parallel (Figure 1B). The replication kinetics of Hofbauer cells infected with HCMV prior to HIV-1 infection was significantly higher at 8 and 12 days postinfection than that for cells infected with HCMV following HIV-1 (Figure 1B). In addition, we monitored HIV-1 viral gene transcription 6 days after infection, using real-time polymerase chain reaction (PCR). We noted that cells infected with HCMV prior to HIV-1 exhibited mean fold increases (± standard error [SE]) of 26.4 ± 5.6 in gag mRNA (average Ct values) and 16.9 ± 2.9 in env mRNA, compared with cells infected with HIV-1 alone. Cells infected with HCMV following HIV-1 exhibited mean fold increases (±SE) of 12.6 ± 1.6 in gag mRNA (average Ct values) and 4.12 ± 0.54 in env mRNA, compared with HCMV-naive cells (Figure 1C). These results demonstrate that HCMV infection of Hofbauer cells render these cells more susceptible to HIV-1 infection in vitro.

Figure 1.

Human cytomegalovirus (HCMV) coinfection increases human immunodeficiency virus type 1 (HIV-1) susceptibility and replication in placental Hofbauer cells. A, Placental Hofbauer cells were infected with HCMV (strain TB40/E-GFP) at a multiplicity of infection of 0.1, or mock infected. Percentages of HCMV-infected cells at 24 and 48 hours postinfection were determined by detection of green fluorescent protein (GFP) (TB40/E-GFP-infected cells) by flow cytometry. Data shown are expressed as the mean ± standard error (SE) of biological triplicates from 10 individual donors analyzed by 1-way analysis of variance (ANOVA) and Dunnett multiple comparison test. **P < .0001. B, Hofbauer cells were infected with HIV BaL strain (HIV-1_BaL) alone or coinfected with HCMV (TB40/E-GFP) 2 days before (gray triangle) or 2 days postinfection (black circle) with HIV-1_BaL. HIV-1 replication was measured in the cell supernatants over time by HIV-1 p24 viral antigen enzyme-linked immunosorbent assay. C, Six days after viral infection, HIV-1- and HCMV/HIV-1–coinfected Hofbauer cells’ messenger RNA levels were measured by real-time polymerase chain reaction to determine the relative expression of gag and env. Gene expression data are represented as fold change relative to time-matched, Hofbauer cells infected with HIV-1 alone (gene expression normalized to β-actin – ΔΔ cycle threshold method). Data shown are expressed as the mean ± SE of biological triplicates from 10 individual donors analyzed by 2-way ANOVA and Sidak multiple comparison test. **P < .0001 and *P < .001 indicate significance between HIV-1–infected cells and coinfected cells. #P < .001 indicates significance among cells coinfected with HCMV 2 days before or 2 days after infection with HIV-1_BaL.

HCMV Infection Upregulates HIV-1 Co-receptor CCR5 Expression in Hofbauer Cells

To understand how HCMV increases HIV-1 susceptibility in Hofbauer cells, we measured expression of the HIV-1 co-receptors CCR5 and CXCR4 following HCMV infection. Analysis by real-time PCR showed a 3-fold increase in CCR5 mRNA following exposure to HCMV, compared to the uninfected cells (Figure 2A). Analysis by flow cytometry showed that HCMV induced an increase in the surface expression of CCR5; however, this increase was not significant (Figure 2B and 2C). Unlike CCR5, CXCR4 displayed a significant decrease in mRNA expression following HCMV infection; however, at the protein level the surface expression remained unchanged (Figure 2A–C).

Figure 2.

Human cytomegalovirus (HCMV) infection upregulates human immunodeficiency virus type 1 coreceptor CCR5 expression in Hofbauer cells. A, Hofbauer cells were infected with HCMV (strain TB40/E-GFP) or mock infected. Twenty-four hours after viral infection, cells’ messenger RNA levels were measured by real-time polymerase chain reaction to determine the relative expression of CCR5 and CXCR4. Gene expression data are represented as fold change relative to time-matched, mock-infected cells (gene expression normalized to β-actin – ΔΔ cycle threshold method). B, Surface expression of CCR5 and CXCR4 was determined by flow cytometry 48 hours postinfection. C, Representative histograms are provided. Data shown are expressed as the mean ± standard error of biological triplicates from 10 individual donors analyzed by 2-way analysis of variance and Sidak multiple comparison test. **P < .0001 and *P < .01 indicate significance between HCMV-infected cells and mock-infected cells.

HCMV Infection Induces Activation of Hofbauer Cells

To investigate the role of HCMV on immune activation placental macrophages, we measured surface expression of the FcγRIII receptor CD16, costimulatory molecules CD80, CD86, and HLA-DR, and the immune regulatory molecule PDL-1, on mock- and HCMV-exposed Hofbauer cells. Consistent with enhanced HIV-1 viral replication, significant up-regulation was noted for CD80 in cells exposed in culture to HCMV as compared to mock-infected cells (Figure 3A). In addition, PDL-1 was also significantly upregulated on cells cultured with HCMV. We noted that CD16 was slightly down-regulated (not significant) on HCMV-infected cells (Figure 3A). Taken together, these findings indicate that Hofbauer cells become immunologically activated following exposure to HCMV.

Figure 3.

Human cytomegalovirus (HCMV) infection induces direct and bystander activation of Hofbauer cells. A, Hofbauer cells were infected with HCMV (strain TB40/E-GFP) or mock infected. Surface-marker expression of CD16, CD80, CD86, PDL-1, and HLA-DR was determined by flow cytometry 48 hours postinfection (hpi). B, Surface-marker expression and green fluorescent protein (GFP) detection was measured by flow cytometry 48 hpi in total cells from cultures exposed to HCMV (all, gray bars), in HCMV-exposed uninfected bystander cells (GFP-negative, black bars) and in HCMV-infected cells (GFP-positive, white bars), within the same culture. C, Surface expression of CCR5 and CXCR4 was determined by flow cytometry 48 hpi in GFP-negative and positive cells. Data shown are expressed as the mean ± standard error of biological triplicates from 10 individual donors analyzed by 2-way analysis of variance and Sidak multiple comparison test. ***P < .0001, **P < .001, and *P < .05 indicate significance between mock-infected cells and HCMV-infected cells (A) or GFP-negative and GFP-positive cells (B).

HCMV Induces CCR5 Expression in Uninfected Bystander Cells

HCMV infection generates a robust local immune response, which may have a significant effect on uninfected bystander cells. At the MOI and time point of infection used in the previous immunophenotypic analysis, approximately 50% of the Hofbauer cells were HCMV infected (GFP-positive) (Figure 1A). Here we wanted to differentiate the HCMV-exposed, uninfected bystander cells (GFP-negative) and the HCMV-infected cells (GFP-positive). As shown in Figure 3B, CD16 and PDL-1 are significantly downregulated in the cells directly infected with HCMV (GFP-positive) compared to those in HCMV-exposed uninfected bystander cells (GFP-negative). In contrast, CD80 expression on HCMV-infected cells is significantly elevated, compared with little to no expression on uninfected bystander cells. A slight down-regulation of HLA-DR was noted in HCMV-infected cells. Interestingly, CCR5 was expressed at a significantly higher level on uninfected bystander cells, compared with the HCMV-infected cells (Figure 3C). Altogether these data show that in HCMV-exposed cultures, cells directly infected with HCMV have a distinct phenotype compared with exposed-uninfected cells. While direct infection may promote cellular activation, soluble factors released in cell supernatants upon HCMV encounter may be responsible for CCR5 induction and activation of uninfected bystander cells.

Proinflammatory Cytokines and Type I IFNs Are Produced by Hofbauer Cells in Response to HCMV Infection

The placenta is recognized as an immunoquiescent milieu, characterized by the predominance of immunoregulatory cytokines [21, 30], which may support the low rate of in utero transmission during maternal HIV-1 infection. It is well documented that HCMV infection leads to rapid expression of type I IFNs and proinflammatory cytokines [15, 16], which contributes to establishment of an antiviral state in infected and surrounding tissues. To assess the effect of HCMV infection on Hofbauer cells, we measured proinflammatory cytokines (TNF-α and IL-6), IL-10, and type I IFNs (IFN-α and IFN-β) in supernatants from HCMV-infected cells by ELISA. Following HCMV infection, we observed increased secretion of the proinflammatory cytokines TNF-α and IL-6, which have previously been shown to up-regulate HIV-1 replication in Hofbauer cells [21]. Conversely, secretion of the immunoregulatory cytokine IL-10 was reduced to levels nearly below the level of detection (Figure 4A). Interestingly, HCMV infection of Hofbauer cells induced a significant increase in IFN-α and IFN-β production, compared with uninfected controls (Figure 4B). These data further demonstrate that HCMV infection of Hofbauer cells induces an inflammatory immune response.

Figure 4.

Proinflammatory cytokines and type I interferons (IFNs) are produced by Hofbauer cells in response to human cytomegalovirus (HCMV) infection. A and B, Hofbauer cells were infected with HCMV (strain TB40/E-GFP) or mock infected. Forty-eight hours after viral infection, cytokine (tumor necrosis factor alpha [TNF-α], interleukin 6 [IL-6], interleukin 10 [IL-10]) and type I IFN levels in the supernatants were determined by enzyme-linked immunosorbent assay. All values are represented as pg/mL. Data shown are expressed as the mean ± standard error of biological triplicates from 10 individual donors analyzed by 2-way analysis of variance and Sidak multiple comparison test. **P < .0001 and *P < .01 indicate significance between mock-infected and HCMV-infected cells.

HCMV Induces a Robust Antiviral Response in Placental Hofbauer Cells

To characterize the type I IFN response in placental Hofbauer cells, we measured mRNA concentrations of the type I IFNs (IFN-α and IFN-β), along with key type I IFN signaling proteins (RIG-I, MDA-5, Jak1, Jak2, STAT1, and STAT2,) by quantitative PCR. Comparisons were made between mock-infected cells vs cells treated with IFN-α (1000 U/mL) or IFN-β (1000 U/mL), or infected with HCMV for 24 hours. Here we show that placental Hofbauer cells are sensitive to exogenous treatment with the type I IFNs. Cells treated with IFN-α or IFN-β significantly up-regulated the mRNA expression IFN-α by approximately 10-fold (Figure 5A). RIG-I and MDA-5 were also significantly upregulated by IFN-α and IFN-β (Figure 5B). The type I IFNs stimulated increased expression of STAT1 and STAT2 (Figure 5D), while only IFN-β induced a significant increase in Jak2 mRNA (Figure 5C). HCMV infection also induced a robust type I IFN response. HCMV infection of Hofbauer cells significantly upregulated mRNA expression of IFN-α and IFN-β compared to mock-infected and type I IFN–treated cells. HCMV infection also induced increased mRNA expression of RIG-I, MDA-5, and STAT1, similar to cells that are treated with IFN-α or IFN-β alone. However, unlike the type I IFNs, HCMV infection substantially down-regulated STAT2 mRNA levels (Figure 5D). The decrease was significantly lower than that noted in mock-infected and type I IFN–treated cells, suggesting that HCMV may interfere with STAT2 transcription and subsequent signaling.

Figure 5.

Human cytomegalovirus (HCMV) initiates a robust antiviral response in placental Hofbauer cells. Hofbauer cells were infected with HCMV (strain TB40/E-GFP) and treated with interferon alpha (IFN-α; 1000 U/mL) or interferon beta (IFN-β; 1000 U/mL), or left untreated for 24 hours. A–D, The type I IFN response was evaluated by measuring messenger RNA levels of the type I IFNs (IFN-α and IFN-β), along with key type I IFN signaling proteins (RIG-I, MDA-5, Jak1, Jak2, STAT1, and STAT2) by real-time polymerase chain reaction. Gene expression data are represented as fold change relative to time-matched, untreated cells (gene expression normalized to β-actin – ΔΔ cycle threshold method). Data shown are expressed as the mean ± standard error of biological triplicates from 10 individual donors analyzed by 2-way analysis of variance and Sidak multiple comparison test. ****P < .0001, ***P < .001, **P < .01, and *P < .05 indicate significance between untreated cells and HCMV-infected or IFN-treated cells.

HCMV Infection Dampens the Type-I IFN Antiviral Response in Hofbauer Cells by Targeting STAT2 Phosphorylation

Studies have shown that while HCMV induces a rapid type I IFN response, this herpesvirus also has the potential to inhibit the overall antiviral pathway [17, 18]. We observed a significant decrease in STAT2 mRNA expression following HCMV infection (Figure 5D). Compared to the other STAT proteins, STAT2 is an essential mediator of signal transduction specifically through the type I IFN receptor [31]. To determine whether this decrease in mRNA transcription translated to protein expression, we analyzed total protein levels and protein phosphorylation of STAT1 and STAT2 in cells treated with IFN-α, infected with HCMV, compared to untreated Hofbauer cells (Figure 6A). We noted that IFN-α significantly up-regulated the total levels of STAT1 and STAT2, compared to uninfected Hofbauer cells. In HCMV-infected cells, the total level of STAT1 was similar to the IFN-α-treated cells and was significantly higher than in uninfected cells. However, induction of STAT2 by HCMV was slightly lower at 24 hours postinfection, compared to the IFN-α–treated cells. Furthermore, the level of endogenous STAT2 decreased significantly over time. Along with decreases in total STAT2 levels, HCMV infection also induced a significant decrease in phosphorylated STAT2 (pSTAT2). To analyze the Western blot, we measured the relative pixel intensities of pSTAT1 and pSTAT2 in IFN-α–treated cells and HCMV-infected cells (Figure 6B). Similar to IFN-α, HCMV infection induced the protein expression of pSTAT1. However, unlike IFN-α, HCMV infection of Hofbauer cells significantly decreased STAT2 phosphorylation over time. These findings indicate that HCMV infection may dampen or block the type I IFN antiviral response by interfering with STAT2 gene expression, protein production, and subsequent phosphorylation.

Figure 6.

Human cytomegalovirus (HCMV) infection dampens the type I interferon (IFN) antiviral response in Hofbauer cells by targeting STAT2 phosphorylation. Hofbauer cells were infected with HCMV (strain TB40/E-GFP), treated with IFN-α (1000 U/mL), or mock infected for 24, 48, and 72 hours. A, Lysates were prepared and subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis. The gel was blotted and the indicated proteins were detected by immunoblotting with indicated antibodies. B, The pixel intensities for pSTAT1 and pSTAT2 of the IFN-α–treated and HCMV-infected Hofbauer cells were quantified using the Odyssey Imaging system and normalized to the β-actin signal. Data shown are expressed as the mean + standard error of biological triplicates from 6 individual donors analyzed by 2-way analysis of variance and Sidak multiple comparison test. *P < .01 indicates significant changes over time (24 hours vs 48 or 72 hours postinfection [hpi]).

DISCUSSION

HCMV reactivation and viremia occur relatively commonly in pregnant women living with HIV. Studies following a cohort of pregnant women from Botswana and Kenya [32, 33] showed that 24% and 17%, respectively, had systemic HCMV viremia at delivery. Interestingly, recent data report strong associations between maternal HCMV viremia and MTCT of HIV-1 [8, 9, 11]. At the placenta, Hofbauer cells are thought to be key mediators involved in HIV-1 and HCMV transmission to the fetus [22, 23, 34, 35]. This interface has been shown to limit HIV-1 replication and transmission to the fetus [21, 24, 25], yet viral-induced local inflammation and cellular activation with maternal HCMV may facilitate MTCT of HIV-1. These observations underscore the importance of understanding the mechanisms of HIV-1/HCMV coinfection at the placenta.

We previously noted that priming fetal CD4⁺ cells with HCMV antigens influenced proliferation and activation of CCR5⁺ central memory T cells, rendering these more susceptible to HIV-1 in vitro [11]. Consistent with these observations, we demonstrate that HCMV infection of Hofbauer cells upregulates expression of CCR5, along with activation markers in vitro, which can lead to increased HIV-1 susceptibility. We also show that HCMV significantly induces the secretion of TNF-α and IL-6, which can enhance HIV-1 replication. Previous studies have demonstrated that placentae from transmitting mothers exhibit a proinflammatory pattern of cytokine release [30]. Although TNF-α has a pivotal role during parturition [36], overproduction of TNF-α may cause damage to the placental barrier, allowing transfer of HIV-1 [37]. HCMV also significantly increases the production of IL-6 in coinfected women and has been shown to contribute to comorbidities associated with persistent inflammation, including a potential impact on the developing fetal immune system in HIV-exposed but uninfected (HEU) infants [38, 39]. The relevance of HCMV to HEU children is unknown but is critically important to determine given the adverse clinical outcomes being reported in this group [40]. Host defenses afforded through anti-inflammatory cytokines endogenously expressed in the placenta, such as IL-10 and transforming growth factor β [21], likely provide protection from transplacental HIV-1 transmission. However, dual maternal infection with HCMV and HIV-1 may provide a 2-hit mechanism that tips the balance toward loss of placental antiviral control.

HCMV infection at the maternal–fetal interface promotes an inflammatory response accompanied by a type I cytokine signature [35]. However, HCMV has evolved mechanisms to antagonize type I IFN signaling [17, 19, 20]. Here we show that HCMV infection of Hofbauer cells interferes with STAT2 protein function. STAT2 mRNA is down-regulated at the transcription level during HCMV infection. HCMV also controls STAT2 protein levels by proteasomal degradation over time. This degradation of STAT2, which regulates type I IFN signaling, has been shown to be sufficient for blocking the type I IFN response [20]. In human fibroblasts, HCMV interferes directly with STAT2 signaling as well as its expression, to antagonize type I IFN downstream effectors [20]. Although the HCMV gene(s) responsible for STAT2 degradation remain unknown, studies suggest that an early gene is responsible [20, 41, 42]. Specifically, the viral immediate early protein IE72 was shown to interact directly with STAT2 and block association of the activated IFN-stimulated gene factor complex with interferon-sensitive response elements in the nucleus, preventing the up-regulation of important antiviral genes, including interferon-stimulated gene-54 and MxA [20]. The overall effect of this interference of the type I IFN defense pathway is dampening of the intrinsic immune response to viral infection, which may promote placental HIV-1 replication and in utero transmission to the fetus.

An interesting observation was the down-regulation of CD16 in Hofbauer cells infected with HCMV. CD16 (FcγRIII) is the Fc receptor necessary for absorption and transport of maternal immunoglobulin G (IgG) and immune complexes across the stroma of the chorionic villi to the fetal endothelium [43, 44]. During healthy pregnancies, newborn IgG levels usually correlate with concentrations from the mother [45]. However, antibody transport during pregnancy can be affected by a number of maternal conditions, including placental infection [46]. Here we demonstrate a marked decrease in CD16 on HCMV-infected Hofbauer cells. Studies have shown that the amount of IgG transmitted during gestation depends on the amount of cell surface receptors [43]. Low levels of this Fc receptor suggest that HCMV infection can limit overall placental antibody transport. Decreases in transplacental antibody transfer from the mother to the fetus may further increase fetal susceptibility to HIV-1 infection.

We also demonstrate the significant increase of PD-L1 on Hofbauer cells following HCMV exposure. The PD-1/PD-L1 pathway has been shown to be central for immune tolerance and during gestation, this pathway may maintain pregnancy partly by regulating T-regulatory cell development and down-regulating maternal T-cell responses. Therefore, upregulation of PD-L1 upon HCMV exposure at the placenta may be a protective mechanism to limit maternal HCMV-specific T cell–mediated immunity. Further studies are warranted to determine the role of PD-L1 during HCMV infection at the placenta.

In sum, we show that HCMV infection may increase HIV-1 susceptibility in placental Hofbauer cells through CCR5 up-regulation and immune activation. In addition, HCMV may promote HIV-1 replication and in utero transmission through the induction of proinflammatory cytokines, inhibition of the type I IFN antiviral pathway, and down-regulation of the FcγRIII receptor. Identifying and treating coinfections that predispose infants to in utero, peripartum, or postpartum HIV-1 infection can help to further reduce the incidence of HIV-1 among neonates while improving overall maternal and child health.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Author contributions. E. L. J. and R. C. designed the research. E. L. J., S. B., E. S. J., A. L., P. A., and S. K. B. performed experiments. E. L. J., K. M. K., and R. C. contributed to interpretation of results and discussion. E. L. J. and R. C. analyzed the data, prepared figures and prepared the manuscript. E. L. J., K. M. K., and R. C. revised the manuscript. and R. C. is the principal investigator. All authors read and approved the final manuscript.

Acknowledgments. The authors thank all the donors who provided informed consent, and Drs Christian Sinzger (Ulm University, Germany) and Don Diamond (City of Hope, Duarte, California) for providing the HCMV strain TB40/E-GFP.

Financial support. This work is supported by the Emory Center for AIDS Research (grant number P30AI050409).

Potential conflicts of interest. All authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.