Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Apr 14;12(4):e1005547.
doi: 10.1371/journal.ppat.1005547. eCollection 2016 Apr.

PPARγ Is Activated During Congenital Cytomegalovirus Infection and Inhibits Neuronogenesis From Human Neural Stem Cells

Affiliations
Free PMC article

PPARγ Is Activated During Congenital Cytomegalovirus Infection and Inhibits Neuronogenesis From Human Neural Stem Cells

Maude Rolland et al. PLoS Pathog. .
Free PMC article

Abstract

Congenital infection by human cytomegalovirus (HCMV) is a leading cause of permanent sequelae of the central nervous system, including sensorineural deafness, cerebral palsies or devastating neurodevelopmental abnormalities (0.1% of all births). To gain insight on the impact of HCMV on neuronal development, we used both neural stem cells from human embryonic stem cells (NSC) and brain sections from infected fetuses and investigated the outcomes of infection on Peroxisome Proliferator-Activated Receptor gamma (PPARγ), a transcription factor critical in the developing brain. We observed that HCMV infection dramatically impaired the rate of neuronogenesis and strongly increased PPARγ levels and activity. Consistent with these findings, levels of 9-hydroxyoctadecadienoic acid (9-HODE), a known PPARγ agonist, were significantly increased in infected NSCs. Likewise, exposure of uninfected NSCs to 9-HODE recapitulated the effect of infection on PPARγ activity. It also increased the rate of cells expressing the IE antigen in HCMV-infected NSCs. Further, we demonstrated that (1) pharmacological activation of ectopically expressed PPARγ was sufficient to induce impaired neuronogenesis of uninfected NSCs, (2) treatment of uninfected NSCs with 9-HODE impaired NSC differentiation and (3) treatment of HCMV-infected NSCs with the PPARγ inhibitor T0070907 restored a normal rate of differentiation. The role of PPARγ in the disease phenotype was strongly supported by the immunodetection of nuclear PPARγ in brain germinative zones of congenitally infected fetuses (N = 20), but not in control samples. Altogether, our findings reveal a key role for PPARγ in neurogenesis and in the pathophysiology of HCMV congenital infection. They also pave the way to the identification of PPARγ gene targets in the infected brain.

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1
Fig 1. Characterization of neural stem cells from human ES cells (NSCs).
Representative immunofluorescence analyses of NSCs cultured in growth medium (A-H) or in differentiation medium, 8 days after the onset of differentiation (I-L), using DAPI staining (A, E, I), or antibodies specific to Nestin (B), SOX2 (C), βIII tubulin (F, J) and HUC/D (G, K). Merged pictures are shown (D, H, L). In differentiation medium, neurons positive for βIII tubulin (J) and HUC/D (K) went alongside to undifferentiated NSCs, which nuclei appeared blue in the merged picture (L). Scale bar: 50 μm.
Fig 2
Fig 2. NSCs are permissive to HCMV infection.
(A) Immunofluorescence analysis of NSCs infected by live (HCMV) or UV-irradiated (HCMV+UV) HCMV, or uninfected (NI), showing nuclear staining to the HCMV Immediate Early antigen (IE) two days post infection (dpi) at a multiplicity of infection (MOI) of 10. DAPI staining and merged pictures are shown. Scale bar: 50 μm. (B) Top: automated counting of immunofluorescence data showing increasing numbers of IE-positive NSCs over time in cultures infected by live HCMV at a MOI of 1 or 10, but not in cultures infected by UV-irradiated HCMV or in uninfected cultures. Data represent means ± CI of 2 independent experiments, each being performed in triplicate. Bottom: western blot analysis showing increasing levels along time of the 72 and 86 kDa isoforms of IE in infected NSCs (MOI 10). (C) Western blot analysis showing production of the early and late HCMV antigens UL44 and pp28, respectively, in infected NSCs (MOI 10), at 8 days pi. (D) Top: transmission electron microscopy of NSC cultures infected by HCMV (MOI 10), showing a cytomegalic NSC (arrowhead) and lipid vesicles (arrows), close to two morphologically normal NSCs (NI), and HCMV particles adsorbed onto the cell surface (inset). Scale bar: 5μm or 0.2 μm (inset). Bottom: transmission electron microscopy of the cytoplasm of an infected NSC, revealing mature viral particles (arrowheads) and dense bodies (asterisks). Pictures were taken 6 days after infection. Scale bar: 0.5μm. (E) Titration of viral particles present in the supernatants of infected NSCs (MOI 10). Supernatants were harvested at different times pi (horizontal axis) and were titrated on MRC5 fibroblasts. Data represent means ± CI of 2 independent experiments, each being performed in triplicate. Virus strain was AD169 except for panel A (VHL/E).
Fig 3
Fig 3. HCMV infection of NSCs impairs neuronogenic differentiation in vitro.
(A) Representative immunofluorescence analysis of NSCs infected by HCMV at a MOI of 10 (HCMV) and uninfected (NI) NSC cultures, 6 days after the onset of differentiation (i.e., 5 days post infection [dpi]) using a βIII tubulin antibody. Scale bar: 50μm. (B) Western blot analysis of whole lysates from differentiating NSC infected by live HCMV at a MOI of 10 (HCMV) or by UV-irradiated HCMV (UV), or uninfected controls, showing decreased levels of βIII tubulin (βIII tub) in the infected cultures. (C) Automated immunofluorescence analysis of differentiating NSC cultures infected or not by HCMV with an HUC/D antibody (HUC/D+). Data represent means ± CI of 3 independent experiments, each being performed in triplicate. (D) Automated immunofluorescence analysis of differentiating NSC cultures infected or not by HCMV using the cell death marker Image-It Dead (ID). Inset: immunofluorescence analysis using an antibody specific to activated (cleaved) caspase 3 (casp3) of infected (MOI 1, 2 dpi) or uninfected (NI) NSC cultures. HCMV strain was AD169. Data represent means ± CI of 2 independent experiments, each being performed in triplicate. **: p<0.01, ***: p<0.005.
Fig 4
Fig 4. HCMV infection triggers the expression and activity of PPARγ in NSCs.
(A) Immunofluorescence analysis using antibodies specific to IE or PPARγ showing strong nuclear staining of PPARγ in NSCs infected by HCMV at a MOI of 10 (HCMV), as compared to non infected NSC cultures (NI) or NSCs infected with UV-irradiated HCMV (HCMV+UV). In merged pictures, double stained nuclei appear cyan (PPARγ and DAPI) or magenta (IE and DAPI); nuclei stained by DAPI and PPARγ and IE antibodies appear purple. (B) Western blot analysis showing increased levels of PPARγ polypeptide in NSCs infected by HCMV (MOI 10) (HCMV) as compared to the uninfected control (NI), at 2 days post infection (dpi). (C) Q-RTPCR analysis showing increased levels of PPARγ transcript in NSCs infected by HCMV (MOI 10) (HCMV) as compared to the uninfected control (NI, value set to 1), or NSCs infected with UV-irradiated HCMV (HCMV+UV) at 2 dpi. Data represent means ± CI of 2 independent experiments, each being performed in triplicate. (D) Luciferase reporter assays showing non specific (pGL4) or PPARγ dependent (pGL4-PPRE) luciferase activity in uninfected NSCs (NI), uninfected NSCs treated with rosiglitazone (NI+rosi), NSCs infected by live HCMV at a MOI of 10, 2 days pi (HCMV), and NSCs infected in the presence of T0070907 (HCMV+T0). Data represent means ± CI of 3 independent experiments, each being performed in triplicate. (E) Chromatin immunoprecipitation assays using an antibody against K4-trimethylated histone 3 (H3K4triMe) as the positive control or two different antibodies against PPARγ (H100 and A3409A), showing increased occupancy by PPARγ of PPREs within the DLK1 gene in NSCs infected by HCMV, as compared with uninfected NSCs. Shown are the fold change ratio from infected versus uninfected (NI) cells. Data represent means ± CI of 2 independent experiments, each being performed in triplicate. (F) Oil red O staining showing numerous lipid vesicles in infected NSC cultures (MOI 10) (HCMV) as compared to uninfected NSCs (NI). Virus strain was AD169. Scale bar: 50 μm. *: p<0.05; ***: p<0.005.
Fig 5
Fig 5. Soluble mediators from infected NSCs trigger the expression of PPARγ in uninfected NSCs.
(A) Immunofluorescence analysis using antibodies specific to IE or PPARγ showing strong increase in PPARγ levels in NSCs infected by HCMV (HCMV), or in uninfected NSCs treated with supernatants prepared from HCMV-infected NSC cultures (SN-HCMV), as compared to uninfected NSCs treated with standard growth medium (mock), or NSCs treated with supernatants prepared from uninfected NSC cultures (SN-NI). The asterisk points to a representative cell with PPARγ nuclear staining. (B) Immunofluorescence analysis using antibodies specific to nestin or PPARγ showing strong increase in PPARγ levels in uninfected NSCs exposed to lipid extracts purified from the supernatants of HCMV-infected NSCs cultures (L-HCMV), as compared to NSCs exposed to lipid extracts purified from the supernatants of uninfected NSCs cultures (L-NI). Virus strain was AD169, and the MOI was 10 in all cases. Scale bar: 50 μm.
Fig 6
Fig 6. Increased production of PPARγ agonist 9-HODE in infected NSCs.
(A) Immunofluorescence analysis of IE expression, showing that treatment of the HCMV inoculum by the PLA2 inhibitor MAFP impairs IE expression. NI: uninfected cells in medium containing 50 nM MAFP. (B) LC-MS/MS screening of PUFA-derived lipids produced in NSCs infected by live (HCMV) or MAFP-treated (HCMV+MAFP) HCMV, or in uninfected NSCs (NI), showing significant increase in 9-HODE levels in infected NSCs (top left). Amounts in supernatants are expressed in pg/ml, amounts in cell lysates are expressed in pg/mg protein. Data represent means ± CI of a minimum of 5 independent experiments, each being performed in triplicate. *: p<0.05. (C) Immunofluorescence analysis of PPARγ expression in uninfected NSCs showing increased PPARγ levels and nuclear translocation (arrowhead). (D) Luciferase assay showing increased PPARγ activity in NSCs stimulated by 9-HODE. HCMV strain was AD169. Data represent means ± CI of 2 independent experiments, each being performed in triplicate. ***: p<0.005.
Fig 7
Fig 7. HCMV replication is enhanced by 9-HODE treatment.
(A) Representative immunofluorescence results of IE expression in NSCs infected by HCMV at a MOI of 10 in the presence of increasing concentrations of 9-HODE or the vehicle, 48 h pi. (B) Immunofluorescence analysis showing increasing number over time of IE positive NSCs in cultures infected by live HCMV at a MOI of 10 in the presence of increasing concentrations of 9-HODE or the vehicle, 48 h pi. Data represent means ± CI of 2 independent experiments, each being performed in duplicate. NI: uninfected control. (C) Titration of the viral particles in supernatants of MRC5 fibroblast cultures treated beforehand with supernatants of NSCs infected by live HCMV at a MOI of 10 in the presence of 1 μg/ml 9-HODE or the vehicle, 48 h pi, or uninfected NSCs. HCMV strain was AD169. Data represent means ± CI of 2 independent experiments, each being performed in triplicate. *: p< 0.05; **: p< 0.01.
Fig 8
Fig 8. PPARγ activity negatively regulates neuronogenesis from NSCs.
(A) Top: Representative immunofluorescence analysis of uninfected NSCs stably expressing PPARγ (NSC-Pg), using an antibody specific to PPARγ (green) (right). Wild type uninfected NSC cultures (WT) stained with the same PPARγ antibody (green) are shown as a control (left). Scale bar: 50 μm. Bottom: In vitro neurogenesis assay showing significantly decreased number of HUC/D positive neurons generated from NSC-Pg (PPAR) stimulated by 1 μM rosiglitazone as compared to either untreated NSC-Pg or NSC-GFP (GFP) stimulated or not by rosiglitazone. Data represent means ± CI of 2 independent experiments, each being performed in duplicate. (B) In vitro neurogenesis assay showing significantly decreased number of HUC/D positive (HUC/D+) neurons generated from wild type NSC treated by 0.1μg/ml or 0.5 μg/ml 9-HODE, as compared to control NSCs. Data represent means ± CI of 2 independent experiments, each being performed in duplicate. (C) Left: Analysis of the number of IE positive cells in NSCs infected or not by HCMV at a MOI of 1 and cultured in the presence of 10 nM T0070907 (T0) or the vehicle, 4 days pi, showing significantly lower number of IE positive NSCs in cultures with T0. Data are expressed as the ratio relative to the number of IE-positive cells found in untreated, infected NSCs, which was arbitrarily set to 1. Right: In vitro neurogenesis assay showing significantly increased number of HUC/D positive neurons generated from wild type NSC infected by HCMV at a MOI of 1 and treated by 10 nM T0070907, as compared to untreated, infected NSCs. Data are expressed as the ratio relative to the number of HUC/D-positive neurons found in untreated, uninfected NSCs, which was arbitrarily set to 1. Data represent means ± CI of 2 independent experiments, each being performed in duplicate. Virus strain was AD169. *: p< 0.05; **: p< 0.01; ***: p< 0.005.
Fig 9
Fig 9. Nuclear PPARγ expression in germinative zone of HMV-infected human fetal brains.
Shown are representative results of immunohistological staining of brain sections from fetuses infected by HCMV (A-G) or from controls (H, I) using antibodies against PPARγ (A-E; F, left; G-I) or IE (F, right). The reference number of each donor is indicated at the bottom left of each panel. Clinical details are summarized in Table 1. PPARγ positive cells (arrows) are detected in the germinative, periventricular, areas and in ependyma (double arrow) in cases, but not in controls. Insets show the localization of the optical field within the brain sections (arrowheads). Note the nuclear localization of PPARγ (A-G), the presence of PPARγ positive cell islets surrounding one IE positive cell in two fields from serial sections (F) and clusters of PPARγ immunoreactive cells around lesional tissue (G). Scale bar: 50 μm.
Fig 10
Fig 10. Summary of the immunohistological exploration of PPARγ expression in HCMV fetal cases and controls.
Each symbol represents the mean relative numbers of PPARγ immunoreactive nuclei in optical fields (n = 6, magnification: x40) found in the brain germinal zone for each individual. Thin horizontal line indicates the average ratio of PPARγ positive cells found in patients, and thick horizontal lines indicate the corresponding SEM. Symbols indicate gestational age in weeks.
Fig 11
Fig 11. PPARγ expression is not detected in the white matter of HCMV-infected fetal brains.
Shown are representative results of immunohistological staining of brain sections from fetuses infected by HCMV (A, B) or from controls (C, D). The reference number of each donor is indicated at the bottom left of each panel. Clinical details are summarized in Table 1. PPARγ positive cells (arrows) are detected in the vessels but not in the white matter in patients and controls. Insets show the localization of the optical field within the brain sections (arrowheads). Scale bar: 50 μm.
Fig 12
Fig 12. Proposed model for the role of PPARγ during HCMV infection of NSCs.
HCMV particles (HCMV) carry onboarded a cell-derived packaged cPLA2 (oPLA2, dots), which catalyzes the release of linoleic acid (LA) from host membrane phospholipids (PL) upon infection. LA undergoes oxidization driven by 15-lipoxygenase (LOX), which generates 9-HODE. 9-HODE, in turn, increases PPARγ levels. Activated PPARγ dimerizes with RXR to regulate the expression of host and viral genomes, resulting in impaired neuronogenesis in vitro and enhanced viral replication. M: cell membrane, C: cytoplasm, N: nucleus.

Comment in

Similar articles

See all similar articles

Cited by 4 articles

References

    1. Cannon MJ. Congenital cytomegalovirus (CMV) epidemiology and awareness. J Clin Virol. 2009;46 Suppl 4:S6–10. 10.1016/j.jcv.2009.09.002 - DOI - PubMed
    1. Cheeran MC, Lokensgard JR, Schleiss MR. Neuropathogenesis of congenital cytomegalovirus infection: disease mechanisms and prospects for intervention. Clin Microbiol Rev. 2009; 22(1):99–126, Table of Contents. 10.1128/CMR.00023-08 - DOI - PMC - PubMed
    1. Cinque P, Marenzi R, Ceresa D. Cytomegalovirus infections of the nervous system. Intervirology. 1997;40(2–3):85–97. . - PubMed
    1. Gabrielli L, Bonasoni MP, Santini D, Piccirilli G, Chiereghin A, Petrisli E, et al. Congenital cytomegalovirus infection: patterns of fetal brain damage. Clinical Microbiology and Infection. 2012;18(10):E419–E27. 10.1111/j.1469-0691.2012.03983.x - DOI - PubMed
    1. Cannon M, Davis K. Washing our hands of the congenital cytomegalovirus disease epidemic. BMC Public Health. 2005;5(1):70 10.1186/1471-2458-5-70 - DOI - PMC - PubMed

Publication types

MeSH terms

Grant support

This study was financially supported by the Institut de la Santé et de la Recherche Médicale (INSERM) (inserm.fr), Centre National de la Recherche Scientifique (CNRS) (cnrs.fr), Université Toulouse Paul Sabatier (UPS) (univ-tlse3.fr), Association Française contre les Myopathies (AFM-Téléthon) (afm-telethon.fr), Assistance Publique-Hôpitaux de Paris (AP-HP) (aphp.fr), Université Paris Descartes (UPD) (parisdescartes.fr) and Chinese Academy of Sciences (CAS) (english.cas.cn). MR was financially supported by UPS. XL was financially supported by CNRS and CAS. YS was financially supported by AP-HP and UPD. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Feedback