Human Cytomegalovirus Productively Infects Adrenocortical Cells and Induces an Early Cortisol Response
MARTA TREVISAN,’ URSKA MATKOVIC,’ RICCARDO CUSINATO,’ STEFANO TOPPO,2 GIORGIO PALÙ,”* AND LUISA BARZON’
‘Department of Histology, Microbiology, and Medical Biotechnologies, University of Padova, Padova, Italy
2 Department of Biological Chemistry, University of Padova, Padova, Italy
Following our recent findings on the presence of human cytomegalovirus (HCMV) in the normal human adrenal cortex and in adrenocortical tumors, especially in cortisol-secreting tumors, aim of the present study was to investigate the direct effects of HCMV infection on human adrenocortical cells. To this aim, both clinical isolates and laboratory strains of HCMV were used to assess the early effects of infection on human adrenocortical cell morphology, proliferation, gene expression, and steroidogenic function. Both clinical and laboratory HCMV strains could infect and replicate in primary human adrenocortical cell cultures and in adrenocortical carcinoma cell lines, leading to cytopathic changes. Most importantly, in the first hours post-infection (p.i.), adrenocortical cells showed a significant increase of cortisol and estrogen production, paralleled by up-regulation of steroidogenic acute regulatory protein and expression of steroidogenic enzymes involved in the last steps of adrenal steroidogenesis. This effect was probably due to HCMV immediate-early gene expression, since it was most evident in the early phases p.i. and UV-inactivated viral particles did not affect hormone production. Moreover, the effect on steroidogenesis was HCMV specific, since it was not observed after infection with herpes simplex virus. These data suggest that human adrenocortical cells are permissive to HCMV infection and acutely respond to infection with increased cortisol production. An acute glucocorticoid response is typically triggered by infections and is considered to be critical to host defense against pathogens, although, in the case of HCMV infection, it might also enhance viral replication and reactivation from latency. J. Cell. Physiol. 221: 629-641, 2009. @ 2009 Wiley-Liss, Inc.
Human cytomegalovirus (HCMV) infection is very common and usually asymptomatic in healthy children and adults, but may present as a mononucleosis-like syndrome in young adults. Following primary infection, HCMV establishes lifelong persistence in the host due to latent infection of mononuclear hematopoietic progenitor cells, mature monocyte-derived macrophages, and dendritic cells. In subjects with congenital or acquired defects of cellular immunity, primary HCMV infection, or reactivation from latency can cause serious systemic disease or specific organ damage. Data on HCMV infection of human adrenal gland are scarce and limited to reports of adrenalitis and adrenal insufficiency in AIDS patients (Pulakhandam and Dincsoy, 1990; Hoshino et al., 1997; Razzaq et al., 2002). We recently detected HCMV genome and expression of both early and late HCMV genes in the normal adrenal cortex and in human adrenocortical tumors and demonstrated the presence of HCMV was significantly more frequent and at higher titer in cortisol-secreting tumors than in non-functioning tumors (Barzon et al., 2008).
The direct effects of HCMV infection on human adrenocortical cells have never been investigated, whereas some data are available on murine cytomegalovirus (MCMV), which has been shown to productively infect adrenocortical cells in vitro and in vivo (Shanley et al., 1979; Shanley, 1987; Reddehase et al., 1988; Price et al., 1996) and to cause focal necrosis in the zona reticularis of the adrenal cortex (Reddehase et al., 1988). In immunocompetent mice, MCMV infection induces adrenalitis but does not compromise adrenal function, since it leads to increased ACTH and corticosterone levels (Price et al., 1996). This effect of MCMV infection was suggested to be mainly mediated by proinflammatory cytokines, such as interleukin I (IL-1) and IL-6, through activation of the hypothalamus-pituitary-adrenal (HPA) axis (Price et al., 1996; Ruzek et al., 1997). Adrenocortical response was critical to
survival, since adrenalectomized mice had an exaggerated cytokine response to viral infection and died when given low, normally tolerated, viral doses (Price et al., 1996; Ruzek et al., 1999). All these studies in the murine model led to the hypothesis that adrenocortical response to viral infection was mediated by cytokines released by inflammatory cells, which activated the HPA axis (Silverman et al., 2004), but have not addressed the hypothesis of a direct effect of viral infection on the adrenal gland.
The present study investigated whether HCMV infects human adrenocortical cells in vitro and the effects of viral infection on adrenocortical cell function and steroidogenesis. We demonstrated here for the first time that HMCV infects and replicates in human adrenocortical cells leading to cytopathic changes and increased cortisol production due to activation of steroidogenesis.
Giorgio Palù and Luisa Barzon contributed equally to this work.
Contract grant sponsor: Veneto Region, Italy;
Contract grant number: RSF 271/07.
Contract grant sponsor: University of Padova, Italy;
Contract grant number: 60A06-9578/07.
*Correspondence to: Giorgio Palù, Department of Histology, Microbiology, and Medical Biotechnologies, University of Padova, Via A. Gabelli 63, 1-35128 Padova, Italy.
E-mail: giorgio.palu@unipd.it
Received 8 April 2009; Accepted 6 July 2009
Published online in Wiley InterScience (www.interscience.wiley.com.), 17 August 2009. DOI: 10.1002/jcp.21896
Materials and Methods Cell cultures
Human foreskin fibroblasts (HFF), human lung fibroblast MRC-5 cells, and Vero cells (American Type Culture Collection, ATCC, Rockville, MD) were grown in a 5% CO2 atmosphere at 37°℃ in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 100 U/ml penicillin, 100 µg/ml streptomycin (1% P-S). HFF and MRC-5 were used at passages 5-12 and 16-30, respectively. The two human adrenocortical carcinoma cell lines NCI-H295R and SW13 were both obtained from ATCC. NCI-H295R cells were grown in RPMI 1640 medium supplemented with 2% FBS, 1% P-S, and 1% selenium/transferrin/ insulin. SW13 cells were grown in a 0% CO2 atmosphere at 37℃ with Leibovitz medium supplemented with 10% FBS and 1% P-S. All media and supplements were purchased from Invitrogen, Inc. (Carlsbad, CA).
Primary cultures of human adrenocortical carcinoma and adrenocortical adenoma were prepared from fresh biopsies obtained from surgically removed tumor masses, after informed consent by the patients and local ethic committee approval. Briefly, tumor samples were cut into small pieces and digested at 37℃ for 90 min in DMEM medium, containing proteinase K I mg/ml, DNAse 67 mg/ml (both from Sigma-Aldrich, St. Louis, MO) amphotericin B 1%, and P-S 1% (Invitrogen, Inc.). After centrifugation at 1,600 rpm for 10 min, cells were resuspended in DMEM medium containing glutamine 2 mM, amphotericin B 1%, P-S 1%, selenium/transferrin/ insulin 1%, and FBS 10%, and grown at 37℃ in a CO2 atmosphere. All experiments with primary adrenocortical cell cultures were done within 3 days after cell seeding, when adrenocortical cells were the predominant cell type (>90%), and concluded within 48- 72 h, before fibroblast overgrowth.
Virus strains and production of viral stocks
Clinical isolates of HCMV were obtained from urine samples of patients with primary HCMV infection attending our Clinical Microbiology and Virology Unit. After filtration through a 0.2 pm filter, HCMV-positive urine samples were used to infect by spin-inoculation cells grown in shell vial cell cultures. The fibroblast-adapted laboratory strains of HCMV AD169 and Towne (both from ATCC) were expanded, respectively, in HFF and MRC-5 cells, purified from cell lysate and culture medium, titrated by plaque assay, and stored at -80℃ until use. The above protocol was also adapted for the laboratory strains of HSV-I Fand the EGFP-Vp16 V-41 except that these strains were expanded and titrated in Vero cells.
HCMV infection of cells
Infection of adrenocortical cells with HCMV and HSV-1 strains was performed at a multiplicity of infection (MOI) of 2 plaque forming units/cell (pfu/cell), unless elsewhere indicated. For infection, medium was removed and replaced by a low volume of fresh medium containing infectious HCMV or HSV-I at indicated MOI and no FBS. After 2 h incubation at 37℃ for HCMV and I h incubation at 37℃ for HSV-I, infection medium was removed and cells were washed three times to remove residual input infectivity, then appropriate growth medium was added and cells were grown at 37℃ until testing. Mock infection was performed by treating cells in the same way as those infected except for the presence of the virus. To inactivate HCMV by treatment with UV light, the same amount of virus stock used for experiments with infectious viral particles was placed in a UV Stratalinker 2400 (Stratagene, La Jolla, CA) and exposed to the 254-nm light source at 0.12 J/cm2 for 7 min.
Immunofluorescence analysis
Adrenocortical cells and control cells were grown on glass coverslips and infected as above described. Coverslips were harvested, washed twice with phosphate-buffered saline (PBS) and then fixed in cold acetone 90% for 25 min at -20℃ for HCMV proteins and in a mixture of methanol-acetone I:I for 5 min at -20℃ for HSV-I protein. Fixed cells were incubated for 60 min at 37℃ with a primary mouse antibody against the immediate-early antigen of HCMV pp72 (E13 IgGI type, Chemicon@ International, Inc., Temecula, MA; Cat. No. 5026), the phosphoprotein of HCMV matrix pp65 (CINA IgGI type, Chemicon@ International, Inc .; Cat. No. 5097i) or the gB protein of HSV-I (mouse lgG2a(k) type, Virusys Corporation, Sykesville, MD; Cat. No. PI 105). Cells were then washed twice in PBS and labeled for 30 min at 37℃ in the dark with fluorescently secondary goat anti-mouse antibody (Ig-FITC, Cat. No. 5024, Chemicon@ International, Inc.). Images were taken with a fluorescent microscope and camera (Leica Microsystems GmbH, Wetzlar, Germany).
Quantitative real-time RT-PCR analysis of HCMV gB mRNA expression
NCI-H295R and SW13 cells were infected with HCMV AD169 and Towne as above described. At 1, 3, 5, and 7 days post-infection (p.i.), cells were harvested for total RNA purification and retro-transcription and the cDNA obtained was used as template for quantitative real-time RT-PCR analysis of HCMV gB expression, performed on a ABI PRISM 7900 HT Sequence Detection System (Applied Biosystems, Foster City, CA). Sequences of PCR primers used for gB amplification are reported in Table 1.
| Gene | Forward primer | Reverse primer | TaqMan probe | T (℃) |
|---|---|---|---|---|
| HCMV | TCATCCACACTAGGAGAGCAGACT | GCCAAGCGGCCTCTGAT | ACTGGGCAAAGACCTTCATGCAGATCTC | 60 |
| BGL | AGGGCCTCACCACCAACTT | GCACCTGACTCCTGAGGAGAA | ATCCACGTTCACCTTGCCCCACA | 60 |
| HCMV gB | GAGGACAACGAAATCCTGTGGGC | GTCGACGGTGGAGATACTGCTGAGG | a | 68 |
| CYP17 | TCTCTGGGCGGCCTCAA | AGGCGATACCCTTACGGTTGT | TGGCAACTCTAGACATCGCGTCC | 60 |
| CYP2 1 | TCAGGTTCTTCCCCAATCCA | TCCACGATGTGATCCCTCTTC | a | 60 |
| CYP11B1 | GGCAGAGGCAGAGATGCTG | TCTTGGGTTAGTGTCTCCACCTG | TGCTGCACCATGTGCTGAAACACCT | 58 |
| CYP11B2 | GGCAGAGGCAGAGATGCTG | CTTGAGTTAGTGTCTCCAGGA | CTGCACCACGTGCTGAAGCACT | 68 |
| CY11A | TCCAGAAGTATGGCCCGATT | CATCTTCAGGGTCGATGACATAAA | a | 60 |
| CYP19 | TCACTGGCCTTTTTCTCTTGGT | GGGTCCAATTCCCATGCA | a | 60 |
| StAR | CCACCCCTAGCACGTGGA | TCCTGGTCACTGTAGAGAGTCTCTTC | a | 60 |
| HSD2B3 | GCGGCTAATGGGTGGAATCTA | CATTGTTGTTCAGGGCCTCAT | TGATACCTTGTACACTTGTGCGTTAAGACCCA | 60 |
| DAX1 | CCAAGGAGTACGCCTACCTCAA | ACTGGAGTCCCTGAATGTACTTCC | a | 60 |
| LRH-1 | TACCGACAAGTGGTACATGGAA | CGGCTTGTGATGCTATTATGGA | a | 60 |
| SF1 | GGAGTTTGTCTGCCTCAAGTTCA | CGTTCTTTCACCAGGATGTGGTT | a | 60 |
| SHP | CCTCAATGCTGTCTGGAGTCCTT | CTGCAGGTGCCCAATGTG | a | 60 |
| GAPDH | GAA GGT GAA GGT CGG AGT C | GAA GAT GGT GAT GGG ATT TC | CAAGCTTCCCGTTCTCAGCC | 60 |
aReal-time RT-PCR was performed with SYBR Green I reagents.
Evaluation of cytopathic effect
For analysis of cytopathic effect (CPE), NCI-H295R and SW-13 cells were infected with HCMV (AD169, Towne, and clinical isolates) or HSV-I (F and V-41) and changes in cell morphology were evaluated by phase contrast light microscopy. The negative control was represented by mock-infected cells. The experiment was conducted three times in triplicate.
Apoptosis assay
NCI-H295R cells and SW-13 were seeded in 24 wells plates and infected with viruses at MOI 2 pfu/cell. Etoposide (Sigma-Aldrich) was added to growth medium at 100 p.M concentration as positive control of apoptosis induction. At 1, 3, 7, 11, and 14 days p.i., cells were harvested for annexin-V-FITC/propidium iodide staining by using the Vybrant Apoptosis Assay Kit (Molecular Probes®, Invitrogen, Ltd., Paisley, UK) following the manufacturer’s protocol. Percentages of cells stained with FITC or propidium iodide were determined on a FACScan (Becton-Dickinson, Franklin Lakes, NJ). The experiment was conducted three times in duplicate.
Cell cycle analysis
Subconfluent NCI-H295R and SW-13 cells, seeded in 25 cm2 flasks, were synchronized by incubation with serum-free medium for 24 h before viral infection. At 24 and 72 h p.i., cells were harvested, suspended in 70% ethanol and stored at -20℃ for 30 min for fixation. Fixed cells were then washed in phosphate citrate buffer and in PBS, treated with 0.1 mg/ml RNase (Sigma-Aldrich) for 30 min, and stained with 0.1 mg/ml propidium iodide (Sigma-Aldrich) at room temperature for 30 min. Propidium iodide-positive staining was evaluated in a Becton-Dickinson FACScan and the fraction of sub-GI, GoGI, S, and G2M cells was determined using CellQuest 3.2 software. The experiment was conducted three times in duplicate.
Proliferation assays
NCI-H295R and SW-13 cells were seeded into 96-well plates at 5,000 and 3,000 cells/well density, respectively. After 24 h, cells were infected with HCMV AD169, Towne, or herpes simplex virus type I (HSV-1), F or V-41 at MOI 2 pfu/cell. At 1, 3, 7, 11, and 14 days p.i. for HCMV and 8, 24, 48, and 72 h p.i. for HSV-I, cell viability was assessed by using the MTT (Sigma-Aldrich) assay, according to the manufacturers’ instructions. The experiments were conducted two times in quadruplicate.
HCMV replication kinetics
MRC-5, HFF, NCI-H295R, and SW-13 cells were infected with HCMV AD169 and Towne at MOI 0.1 pfu/cell. At 1, 3, 5, 7, days p.i., cells and culture medium were collected for HCMV titration by plaque assay on HFF cells and for HCMV DNA quantification by quantitative real-time PCR after DNA purification with the QIAamp® DNA mini kit (QIAGEN GmbH, Hilden, Germany). The real-time PCR method for quantitative analysis of HCMV DNA has been previously described (Mengoli et al., 2004). AD169 and Towne replication kinetics was also evaluated in the presence or absence of hydrocortisone and aminoglutethimide. In particular, NCI-H295R, SW-13, and HFF cells were pre-treated with hydrocortisone 10 p.M (Solu-Cortef®, Pharmacia, Milano, Italy) or aminoglutethimide 300 µM ((S)-(-) aminoglutethimide, Sigma- Aldrich) during the 48 h before AD169 infection. Then, AD169 infection was carried out at MOI 0.1 pfu/cell in the presence of hydrocortisone 10 p.M and aminoglutethimide 300 M, alone or in combination. At 1, 3, 5, 7, 9, and 11 days p.i., cells and culture media were collected for measurement of HCMV DNA load by quantitative real-time PCR.
Evaluation of adrenal steroid hormone production
Subconfluent NCI-H295R cells were infected with HCMV (AD169 and Towne) or HSV-I (F and V-41). At different time points p.i., culture media were harvested for steroid hormone measurements. Cortisol, 17ß-estradiol, and aldosterone were measured in culture medium by EIA assays (Cayman@ Chemical, Ann Arbor, MI); DHEAS was measured by an ELISA assay (DHEAS Direct ELISA Kit, dbc-Diagnostics Biochem Canada, Inc., London, ON, Canada) according to the manufacturer’s instructions.
Quantitative real-time RT-PCR analysis of steroidogenic enzymes expression
Subconfluent NCI-H295R cells were infected with HCMV and HSV-1 as above described. Cells were harvested for total RNA purification and retro-transcription and the cDNA obtained was used as template for quantitative real-time RT-PCR analysis, performed on a ABI PRISM 7900 HT Sequence Detection System (Applied Biosystems). Investigated genes were: CYPI7 (17a-hydroxylase), CYP21 (21ß-hydroxylase), CYPI IBI (11ß-hydroxylase), CYPI IB2 (aldosterone synthase), CYPI IA (cholesterol side-chain cleavage), CYP19 (aromatase), StAR (steroidogenic acute regulatory protein), 3 B-HSD2 (3ß-hydroxysteroid dehydrogenase), DAXI (dosage-sensitive sex reversal, adrenal hypoplasia congenita critical region on the X chromosome, gene-1), LRH-1 (liver receptor homolog-1), SFI (steroidogenic factor-1), SHP (small heterodimer partner). For data normalization, expression of the housekeeping gene GAPDH was evaluated. Sequences of primers and probes are reported in Table 1.
Microarrays analysis
Total RNA extracted from NCI-H295R cells infected by AD 169 at MOI 2 pfu/cell was amplified by using the SuperScript™ Indirect RNA Amplification System (Invitrogen, Inc.) following manufacture’s instructions. The aRNA was labeled with CyTM3 and CyTM5 (Amersham CyScribe™M Post-Labeling Kit, Amersham, Little Chalfont, Buckinghamshire, UK) for 2 h at room temperature in the dark and hybridized in a Tecan HS 400™M hybridization station (Tecan Group Ltd, Männedorf, Switzerland) for 16 h at 39°℃ to DNA microarray slides containing 21,329 spotted oligonucleotide sequences of human genes (CRIBI, University of Padova, Italy). Microarray experiments were performed in duplicate including dye swap. The arrays were scanned using an Affymetrix 428 scanner (MWG Biotech, Ebersberg, Germany) and the images analyzed by ImaGene and GeneSight softwares (BioDiscovery, Inc., El Segundo, CA). Data were filtered by concordance between the two internal replicates. This was done by evaluating the compatibility of values for each point of time taken individually and then considering them together.
Functional annotation assessment and classification
To enrich functional annotations of modulated cellular genes, an in silico prediction and subsequent manual validation of the results integrated by scientific literature was applied. Function prediction was carried out using the recently developed ARGOT tool (Fontana et al., 2009) able to extract significant gene ontology (GO) (Ashburner et al., 2000) terms from BLAST (Altschul et al., 1997) searches versus UniProt Databank (Apweiler et al., 2004) by means of semantic similarities. The GO annotated sequences were further mapped into generic GO functional categories. We obtained the classification reported in Table 2 mapping GO annotated sequences into the official Go Slim available at the gene ontology web site (http://www.geneontology.org/ GO.slims.shtml) and performed further manual check of the obtained functional classes.
| Official symbol | Full gene name | GB accession | Time of max change (h p.i.) | Log fold change |
|---|---|---|---|---|
| Biological process | ||||
| GEFT | RhoA/RAC/CDC42 exchange factor | BC012860 | 8 | 1.17 |
| ZCCHC6 | Zinc finger, CCHC domain containing 6 | AB051498 | 8 | 1.12 |
| RPN2 | Ribophorin II | NM_002951 | 8 | 0.99 |
| PDE12 | Phosphodiesterase 12 | AK001526 | 8 | -1.23 |
| SLC25A3 | Solute carrier family 25 member 3 | AK057575 | 24 | 1.07 |
| NDUFV3 | NADH dehydrogenase (ubiquinone) flavoprotein 3 | NM_021075 | 72 | -1.11 |
| Biosynthetic process | ||||
| RNPEP | Arginyl aminopeptidase (aminopeptidase B) | NM_020216 | 8 | -1.16 |
| STARD4 | StAR-related lipid transfer domain containing 4 | AK054566 | 8 | -1.15 |
| Carbohydrate metabolic process | ||||
| MDH1B | Malate dehydrogenase IB, NAD (soluble) | AK058070 | 8 | -1.45 |
| TALDO1 | Transaldolase I | NM_006755 | 24 | 1.65 |
| Cell adhesion | ||||
| ESAM | Endothelial cell adhesion molecule | AF361746 | 8 | -1.48 |
| CDH6 | Cadherin 6, type 2, K-cadherin (fetal kidney) | NM_004932 | 24 | -1.08 |
| Cell communication | ||||
| PSEN1 | Presenilin 1 (Alzheimer disease 3) | NM_007318 | 8 | 0.98 |
| LRRK2 | Leucine-rich repeat kinase 2 | AK026776 | 8 | -1.60 |
| STARD8 | StAR-related lipid transfer (START) domain containing 8 | NM_014725 | 8 | -1.16 |
| STAT3 | Signal transducer and activator of transcription 3 | NM_003150 | 24 | 1.55 |
| PTGER3 | Prostaglandin E receptor 3 (subtype EP3) | AL050227 | 24 | 1.05 |
| ITGA7 | Integrin, alpha 7 | NM_002206 | 24 | -1.09 |
| SMAP1 | Stromal membrane-associated GTPase-activating protein I | NM_021940 | 72 | 1.39 |
| P2RY5 | P2RY5 purinergic receptor P2Y, G-protein coupled, 5 | NM_005767 | 72 | 1.31 |
| C9orf86 | Chromosome 9 open reading frame 86 | NM_024718 | 72 | 1.12 |
| GOLSYN | Golgi-localized protein | NM_017786 | 72 | 1.06 |
| ARHGAP1 | Rho GTPase activating protein I | BC018118 | 72 | 0.97 |
| ADAM17 | A disintegrin and metalloproteinase domain 17 | NM_003183 | 72 | -1.44 |
| Cell cycle | ||||
| E2F7 | E2F transcription factor 7 | BC016658 | 24 | -0.99 |
| Cell death | ||||
| CANX | Calnexin | NM_001746 | 8 | -2.09 |
| LGALS 1 2 | Lectin, galactoside-binding, soluble, 12 | NM_033101 | 8 | -1.07 |
| TNFAIP8 | Tumor necrosis factor, alpha-induced protein 8 | NM_014350 | 72 | -1.54 |
| Cell differentiation | ||||
| TWIST1 | Twist homolog 1 (Drosophila) | NM_000474 | 8 | 0.96 |
| Cell proliferation | ||||
| TSPAN3 1 | Tetraspanin 31 | NM_005981 | 8 | -1.13 |
| DUSP8 | Dual specificity phosphatase 8 | NM_004420 | 8 | -1.12 |
| IGFBP7 | Insulin-like growth factor binding protein 7 | NM_001553 | 24 | 2.13 |
| CTBP1 | C-terminal binding protein I | AL137653 | 24 | 1.46 |
| Cellular amino acid and derivative metabolic process | ||||
| ASNSD1 | Asparagine synthetase domain containing I | NM_019048 | 24 | 1.64 |
| EMG1 | EMGI nucleolar protein homolog (S. cerevisiae) | NM_006331 | 8 | -1.53 |
| GEMIN4 | Gem (nuclear organelle)-associated protein 4 | AF177341 | 8 | -1.40 |
| Immune system process | ||||
| CD163L1 | CD163 molecule-like I | NM_033330 | 8 | -1.46 |
| PSMB10 | Proteasome (prosome, macropain) subunit, beta type, 10 | NM_002801 | 24 | 1.13 |
| PF4V1 | Platelet factor 4 variant I | NM_002620 | 24 | 1.09 |
| HLA-DQA1 | Major histocompatibility complex, class II, DQ alpha 1 | NM_002122 | 24 | -1.18 |
| PON2 | Paraoxonase 2 | NM_000305 | 72 | -1.21 |
| Multicellular organismal development | ||||
| DISP1 | DISPI dispatched homolog I (Drosophila) | AK056569 | 8 | 1.31 |
| TMEM176B | Transmembrane protein 176B | NM_014020 | 8 | -1.28 |
| TBX1 | T-box 1 | NM_005992 | 24 | -0.96 |
| Nucleobase, nucleoside, | nucleotide, and nucleic acid metabolic process | |||
| DICER1 | Dicer 1, ribonuclease type III | AF007142 | 8 | -1.61 |
| ATP5I | ATP synthase, mitochondrial FO complex | NM_007100 | 8 | -1.02 |
| TOP3A | Topoisomerase (DNA) III alpha | NM_004618 | 24 | 1.57 |
| GINS3 | GINS complex subunit 3 (Psf3 homolog) | NM_022770 | 24 | -1.12 |
| Protein metabolic process | ||||
| DBT | Dihydrolipoamide branched chain transacylase E2 | AK024946 | 8 | -1.05 |
| EHD3 | EH-domain containing 3 | NM_014600 | 24 | 1.01 |
| HUWE1 | HECT, UBA, and WWE domain containing I | AF161390 | 24 | 0.99 |
| FBXO10 | F-box protein 10 | BC013747 | 72 | 0.95 |
| Response to stress | ||||
| CCL20 | Chemokine (C-C motif) ligand 20 | NM_004591 | 24 | 1.09 |
| Transcription | ||||
| DHX9 | DEAD/H box polypeptide 9 (RNA helicase A) | NM_001357 | 8 | 1.53 |
| LCORL | Ligand-dependent nuclear receptor corepressor-like | AL133031 | 8 | 1.04 |
| SFMBT2 | Scm-like with four mbt domains 2 | AB046837 | 8 | 0.99 |
| NSD1 | Nuclear receptor binding SET domain protein 1 | AK001546 | 8 | -1.56 |
| FRYL | FRY-like | AB020633 | 8 | -1.42 |
| ZNF626 | Zinc finger protein 626 | BC007116 | 8 | -1.22 |
| Official symbol | Full gene name | GB accession | Time of max change (h p.i.) | Log fold change |
|---|---|---|---|---|
| ETV1 | Ets variant gene 1 | NM_004956 | 8 | -1.20 |
| NR3C1 | Nuclear receptor subfamily 3, group C, member I | U25029 | 24 | 1.42 |
| SSRP1 | Structure-specific recognition protein 1 | NM_003146 | 24 | 1.26 |
| TWIST1 | Twist homolog I (Drosophila) | NM_000474 | 24 | 1.25 |
| CCNC | Cyclin C | NM_005190 | 24 | 1.08 |
| ZNF187 | Zinc finger protein 187 | BC013962 | 24 | -1.57 |
| ZNF516 | Zinc finger protein 516 | NM_014643 | 24 | -1.08 |
| FOXE1 | Forkhead box EI (thyroid transcription factor 2) | NM_004473 | 72 | 1.35 |
| SNW1 | SNW domain containing I | NM_012245 | 72 | 1.18 |
| PTMA | Prothymosin, alpha (gene sequence 28) | NM_002823 | 72 | 1.09 |
| EEPD1 | Endonuclease/exonuclease/phosphatase family domain containing I | AK027386 | 72 | 0.98 |
| Transport | ||||
| ARFGAP2 | ADP-ribosylation factor GTPase activating protein 2 | NM_032389 | 8 | -1.27 |
| ABCG8 | ATP-binding cassette, subfamily G, member 8 | NM_022437 | 24 | 1.23 |
| SSR3 | Signal sequence receptor, gamma | NM_007107 | 24 | 1.12 |
| SLCO5A1 | Solute carrier organic anion transporter family, 5Al | NM_030958 | 24 | -1.71 |
| TRAM2 | Translocation-associated membrane protein 2 | NM_012288 | 72 | -1.49 |
Genes are grouped according to their ontology.
Statistical analysis
Results are presented as mean ± SD. Comparisons between groups were performed by two-sided unpaired Student’s t-test. Statistical significance was considered at P < 0.05.
Results
HCMV infects and replicates in human adrenocortical cells
To determine whether human adrenocortical cells are permissive to HCMV infection, the human adrenocortical carcinoma cell lines NCI-H295R and SW-13, as well as primary cell cultures established from human adrenocortical carcinomas and adenomas, were infected with three HCMV clinical isolates and with the laboratory HCMV strains AD169 and Towne. While SW-13 cells are derived from a poorly differentiated carcinoma and display malignant phenotype, NCI-H295R cells, besides having a slow growth rate, retain most physiological properties of normal adrenocortical cells, such as intact steroidogenetic pathways and responsiveness to hormone stimuli, so they are considered a good model to investigate adrenocortical cell physiology (Rainey et al., 2004). For infection experiments, HCMV clinical isolates obtained from urine samples were directly used to infect cells, without passages in fibroblasts, whereas AD169 and Towne, after expansion in HFF cells, were used at MOI 2 pfu/cell. At different time points p.i., expression of the HCMV immediate-early antigen pp72 and late antigen pp65 was evaluated by immunofluorescence analysis. As demonstrated in Figure I, positive nuclear pp72 and pp65 immunostaining was detectable in human adrenocortical cells after infection with both HCMV clinical isolates and the laboratory strains. The efficiency of infection was, however, markedly lower than in human lung fibroblasts MRC-5 cells and in HFF cells, which were used as control. The number of HCMV-infected adrenocortical cells increased with time p.i. (Fig. 1B), suggesting productive viral replication. Productive HCMV replication was also suggested by increasing expression of the late HCMV gB gene with time p.i. (Fig. IC).
HCMV has cytopathic effects on human adrenocortical cells
To evaluate whether HCMV infection induced any morphological change in NCI-H295R and SW-13 cells, cells were subcultured several times and observed at different time points p.i. with AD169 and Towne at MOI 2 pfu/cell. As documented by phase contrast micrographs (Fig. 2A),
cytopathic changes appeared at about 7 days p.i. and were characterized by shrinking and rounding of cells, which underwent death with complete lysis of cell monolayer at about 3 weeks p.i. Plaques of cell lysis were evident after Towne infection (Fig. 2A).
The effect of infection on cell viability, apoptosis induction, and cell cycle progression was analyzed during both early (2 and 3 days p.i.) and late (7, 11, and 14 days p.i.) stages of infection. The MTT assay showed that infection with AD 169 and Towne at MOI 2 pfu/cell caused a progressive decrease of cell viability, in agreement with the observed CPE (Fig. 2B). The propidium iodide/annexin-V apoptosis assay, which was used to evaluate the mechanism of death in HCMV infected NCI-H295R cells, demonstrated a slight increase of apoptotic (13% in HCMV- infected cells vs. 3% in mock-infected cells) and necrotic (10% in HCMV-infected cells vs. 2% in mock-infected cells) cells already at 24 h p.i. and a progressive increase of necrotic and apoptotic cells during the time course p.i. (data not shown). Cell cycle analysis of infected NCI-H295R cells showed an accumulation of cells in the sub-Go/GI phase of the cell cycle both at early (75% in HCMV-infected cells vs. 69% in mock-infected cells at 24 h p.i.) and late (76% in HCMV-infected cells vs. 62% in mock-infected cells at 72 h p.i.) stages of infection (data not shown).
Kinetics of HCMV replication in human adrenocortical cells and effect of steroid hormones on HCMV replication
Infection with AD169 and Towne at low MOI (0.1 pfu/cell) was done to asses virus replication kinetics in adrenocortical carcinoma cell lines as compared with the kinetics in control MRC-5 and HFF cells, which are fully permissive to HCMV infection. To determine viral replication kinetics, viral load was measured in culture media and in cell lysates by standard plaque assay and by quantitative real-time PCR. These experiments showed that the efficiency of HCMV infection and replication in NCI-H295R cells was higher than in SW-13 cells, but markedly lower than in MRC-5 and in HFF cells (Fig. 3).
We wondered whether the different permissiveness of the adrenocortical cell lines to HCMV replication was related to the ability of NCI-H295R to produce cortisol and other steroid hormones. Indeed, experimental studies from the literature demonstrate that treatment of human fibroblasts and embryonic kidney cells with pharmacological concentrations of glucocorticoids, such as hydrocortisone (i.e., cortisol) and dexamethasone, enhances HCMV replication, whereas treatment with estrogens and androgens does not (Tanaka
A
a
b
B NCI-H295R
8 hpi
C
d
24 hpi
C
2.5
72 hpi
2
NCI-H295R
HCMV gB mRNA
SW13
1.5
1
T
HFF
0.5
72 hpi
0
1
3
5
7
Time (days pi)
et al., 1984; Koment, 1989). To investigate whether HCMV replication in adrenocortical cells is modulated by steroid hormones, NCI-H295R and SW-13 adrenocortical cells were treated with hydrocortisone and L-aminoglutethimide, an inhibitor of several adrenal steroidogenic enzymes, in a time course experiment of infection with AD 169 at MOI 0. I pfu/cell. MRC-5 and HFF fibroblasts were used as controls. Treatment with hydrocortisone 10 p.M enhanced AD169 replication in fibroblasts leading to a 50% increase of cell-associated and cell- free viral DNA load. In adrenocortical cells, treatment with hydrocortisone had no significant effect on cell-associated AD169 DNA load, but led to peaks of viral DNA in the culture medium of both NCI-H295R and SW-13 cells, suggesting increased virus release. Treatment of adrenocortical cells with L-aminoglutethimide 300 uM, either alone or in combination with hydrocortisone 10 uM, had no significant effect on viral replication (Fig. 4).
Microarray analysis of gene expression profile in AD169-infected NCI-H295R cells
HCMV infection modulates many genes involved in several cellular pathways, as shown by microarray analysis of gene
expression in HCMV-infected fibroblasts and myeloid progenitor cells (Zhu et al., 1998; Browne et al., 2001; Abate et al., 2004; Slobedman et al., 2004; Chan et al., 2008). We performed DNA microarray analysis to get a picture of the global effects of HCMV infection on NCI-H295R gene expression profile. For this purpose, a time-course experiment of infection with AD 169 at MOI 2 pfu/cell versus mock infection was performed, in which NCI-H295R cells were harvested at 8, 24, and 72 h p.i. for gene expression analysis. Based on the results of the analysis of HCMV replication kinetics, these time points correspond to relatively early phases in the HCMV replication cycle. Raw data of the experiments are posted at http://mbi.bio.unipd.it/hcmv/. Cellular genes that were up-regulated or down-regulated twofold or higher upon HCMV infection were identified by the differential regulation analysis tool of the GeneSight software, which was applied at 99% confidence level to determine the significance of average change in expression (ratio of HCMV-infected/mock-infected cells). The functional classification of the modulated cellular genes was carried out using their GO annotations (Ashburner et al., 2000); the list of genes which were significantly up-regulated or down-regulated in HCMV-infected cells during all experimental time points is reported in Table 2. At 8 h p.i.,
A Towne
AD169
mock
B
140
Towne
NCI-H295R cell viability (% of mock infection)
120
-0-AD169
100
80
60
40
20
0
0
2
4
6
8
10
12
14
Time (days pi)
HCMV-infected NCI-H295R cells showed induction of several genes involved in transcription, while genes involved in biosynthetic processes where inhibited. At 24 h p.i., induced genes encoded factors involved in the anti-microbial and stress response and inhibitors of cell proliferation, while genes involved in cell cycle progression where inhibited. At 72 h p.i., HCMV infection of adrenocortical cells mainly induced genes involved in cell communication and in RNA transcription. Interestingly, expression of NR3CI, which encodes the a subunit of the glucocorticoid receptor (GRa) was significantly increased at all time points p.i. Among genes mainly down- regulated at 8 h p.i., there were STARD4 and STARD8, which are involved in cholesterol transport and are homologous to StAR (Table 2). Microarray results on expression data of a subgroup of genes were confirmed by real-time RT-PCR analysis (data not shown).
HCMV infection of human adrenocortical cells induces adrenal steroidogenesis
The observation that HCMV infection of NCI-H295R cells altered expression of genes involved in cholesterol transport and metabolism led us to investigate the effects of infection on steroidogenesis. To analyze steroid hormone response to HCMV infection, subconfluent NCI-H295R cells were infected with AD 169 and Towne at MOI 2 pfu/cell and grown in
AD169
8
MRC5
7
HFF
Log10 pfu/mL
6
H295R
F
5
SW13
I
4
1
2
I
L
1
0
₮
0
1
2
3
4
5
6
7
Time (day pi)
Towne
8
> MRC5
7
o- HFF
Log10 pfu/mL
₮
6
+ H295R
5
SW13
±
₮
4
±
3
2
I
1
₮
0
0
1
2
3
4
5
6
7
Time (day pi)
serum-free medium. At 8, 24, and 72 h p.i., cells and culture medium were harvested to measure, respectively, steroidogenic enzyme gene expression and the levels of the main steroid hormones of the mineralocorticoid (i.e., aldosterone), glucocorticoid (i.e., cortisol), and androgenic (i.e., DHEA-S) pathways, as well as 17ß-estradiol, which is also produced by this cell line. Measurement of steroid hormones after HCMV infection demonstrated a significant increase of cortisol and 17-estradiol production as compared with mock infection and this effect was already apparent at 8 h p.i. (Fig. 5).
Several studies in the literature have suggested that most of the early effects observed in cells after HCMV infection could be due to interaction of the viral particle with host cell surface receptors (Zhu et al., 1997, 1998; Simmen et al., 2001; Boehme et al., 2004). To investigate whether such a rapid steroid hormone response of NCI-H295R cells to HCMV infection was simply a consequence of the interaction of structural components of the virus with infected cells, UV-inactivated viral particles were used for infection and compared with replication-competent virus. When UV-inactivated HCMV was used, no effects on steroid hormone production were observed, thus, indicating that induction of steroidogenesis represented a response to expression of early viral genes and/ or virus replication (Fig. 5). Cells infected with UV-inactivated HCMV did not show any pp72 immunostaining at immunofluorescence analysis, thus, demonstrating that our UV
NCI-H295R, cell associated
NCI-H295R, cell free
50
HC
60000
40
AG
50000
AG-HC
30
40000
Ctrl
30000
20
20000
10
10000
0
0
0
1
2
3
4
5
6
7
8
9
10
11
0
1
2
3
4
5
6
7
8
9
10
11
HCMV DNA copies/cell
SW13, cell associated
20
HCMV DNA copies/mL
30000
SW13, cell free
25000
15
20000
10
15000
10000
5
5000
0
0
0
1
2
3
4
5
6
7
8
9
10
11
0
1
2
3
4
5
6
7
8
9
10
11
1000
HFF, cell associated
HFF, cell free
HC
12000000
5000
Ctrl
9000000
1000
6000000
5000
3000000
0
0
0
1
2
3
4
5
6
7
8
9
10
11
0
1
2
3
4
5
6
7
8
9
10
11
Days post infection
Days post infection
irradiation protocol effectively blocked viral gene expression (data not shown).
To expand this observation on the effect of HCMV on steroidogenesis to other human adrenocortical cell systems, primary cell cultures were obtained from both cortisol-producing and non-functioning human adrenocortical carcinomas and infected with AD169 at MOI 2 pfu/cell. Also in these primary cell cultures, AD169 infection led to increased cortisol production (Fig. 6A).
To elucidate the mechanism of increased steroid hormone production following HCMV infection, we investigated by quantitative real-time RT-PCR the expression profile of genes encoding steroidogenic enzymes and factors involved in steroid hormone synthesis in RNA purified from NCI-H295R cells of the above experiments (i.e., at 8, 24, and 72 h p.i.). This analysis showed that HCMV infection had a marked effect on steroidogenic enzymes expression. In particular, in the early phases p.i. (8 and 24 h p.i.), mRNA levels of StAR were markedly increased (Fig. 7) and this effect was observed also in primary adrenocortical carcinoma cell cultures after HCMV infection (Fig. 6B). StAR is expressed in steroidogenic cells, where it regulates the rate-limiting step of steroidogenesis, that is, cholesterol transport from the outer to the inner mitochondrial membrane where it is converted into pregnenolone by CYP11A. HCMV infection also led to a sharp
increase of CYP11B1 and CYP19 mRNA, which encode, respectively, I IB-hydroxylase and aromatase, that are the key enzymes for cortisol and estrogens synthesis. CYPI IB2, which encodes aldosterone synthase, was up-regulated at later time points p.i. (Fig. 7). Expression of other steroidogenic enzymes, that is, CYPI IA, HSD3B2, and CYP17, was down-regulated after HCMV infection (Fig. 7A). Infection with UV-inactivated viral particles had no effect on steroidogenic enzyme expression.
To get clues on the mechanism of HCMV-mediated modulation of adrenal steroidogenesis, we analyzed expression of nuclear hormone receptors involved in the regulation of steroidogenic enzymes (i.e., SF-I, DAX-I, LRH-I, and SHP) in HCMV-infected cells. SF-I and LRH-I share structural features and activating properties on steroidogenesis through induction of StAR and steroidogenic enzyme expression, whereas DAX-I and SHP inhibit other nuclear receptor activity, especially SF-I and LRH-I activity, respectively, through heterodimeric interactions. During the time course of infection, SF-I and DAX-I mRNA expression was inhibited, while SHP mRNA levels were significantly up-regulated. Modulation of these nuclear hormone receptors was more pronounced in early phases p.i. (8 and 24 h p.i.) than in a later phase at 72 h p.i. (Fig. 7A). LRH-I mRNA levels were very low in NCI-H295R cells and not significantly modified after HCMV infection (data not shown).
800
Mock
*
5
*
Cortisol (ng/ml)
700
AD169
600
DHEA-S (µg/mL)
4
Towne
500
UV-AD169
3
400
UV-Towne
300
2
*
200
*
T
I
1
T
T
T
T
100
T
0
0
8
24
72
8
24
72
Time (hpi)
Time (hpi)
4000
17B-estradiol (ng/ml)
30
*
Aldosterone (pg/mL)
3500
25
3000
*
2500
20
2000
15
1500
*
10
*
1000
T
T
T
500
5
*
T
0
0
8
24
72
8
24
48
Time (hpi)
Time (hpi)
A
200
Mock
4000
*
Mock
Cortisol (nmol/L)
AD169
AD169
*
150
Cortisol (nmol/L)
3000
*
100
2000
50
1000
0
0
ACC9
ACC6R
CPA3
B
30
Mock
30
Mock
*
25
AD169
25
AD169
StAR mRNA
StAR mRNA
20
20
*
15
*
15
10
10
5
5
0
0
ACC9
ACC6R
CPA3
A
10
StAR
0.6
Mock
CYP17
0.25
SF-1
8
AD169
0.5
0.2
6
Towne
0.4
0.15
0.3
4
0.2
0.1
2
0.1
0.05
0
0
0
1.2
CYP11A1
12
0.0025
1
10
CYP21
DAX-1
0.002
0.8
8
0.0015
0.6
6
0.4
4
0.001
0.2
2
0.0005
0
0
0
0.5
0.06
HSD3B2
0.0012
0.05
CYP11B1
SHP
0.4
0.04
0.0009
0.3
0.03
0.0006
0.2
0.02
0.1
0.01
0.0003
0
0
0
0.05
CYP11B2
8
24
72
8
24
72
Time (hpi)
Time (hpi)
0.04
0.03
B
Cholesterol
0.02
STAR
0.01
CYP11A
CYP17
SULT2A1
0
Pregnenolone
17OH Preg
DHEA
DHEA-S
0.6
CYP19
HSD3B2
0.5
0.4
Progesterone
17OH Prog
Androstenedione
0.3
CYP21
HSD17B3
0.2
DOC
Deoxycortisol
Testosterone
0.1
CYP11B2
CYP11B1
CYP19
0
8
24
72
Aldosterone
Cortisol
17ß-estradiol
Time (hpi)
HSV-1 efficiently replicates in NCI-H295R cells but has no effect on steroid hormone production
In order to assess whether the effects of HCMV infection on adrenal steroidogenesis were common to other herpesviruses, the same infection experiments were performed with HSV-I. NCI-H295R cells were at this purpose infected with two laboratory strains of HSV-I (i.e., HSV-I Fand EGFP-vp 16 HSV-I V-41) at MOI 2 pfu/cell. Efficiency of HSV-I infection was assessed by detection of EGFP expression, after infection with the V-41 strain, or by immunostaining against HSV-I gB protein, after infection with the F strain. Both HSV-I laboratory strains demonstrated to be able to infect NCI-H295R cells: 100% of cells resulted positive for V41 EGFP or for F gB expression within 72 h p.i. (data not shown). Cytopathic changes occurred within the first 48 h p.i. and led to rapid disruption of cell monolayer (Fig. 8A). The MTT test and propidium iodide/ annexin-V apoptosis assay were also performed in order to evaluate effect of viral infection on cell viability. These assays showed that both F and V-41 HSV-I strains caused cell death mainly due to apoptosis at the end of virus life cycle (Fig. 8B). The effect of HSV-I infection on adrenal hormone production and steroidogenesis gene expression was evaluated throughout a time course experiment, as above described for HCMV. As shown in Figure 8℃, both HSV-I strains had no significant effect
on steroid hormone production. No effects were also observed on expression of StAR and steroidogenic enzymes (data not shown). Thus, the steroid hormone response appears to be HCMV specific and not a common response to viral infection.
Discussion
In this study, we characterized adrenocortical cell response to HCMV infection. We demonstrated here for the first time that HCMV productively infects cultured human adrenocortical cells, which acutely respond to infection with up-regulation of StAR and steroidogenic enzyme expression and increased cortisol production.
In vivo, during primary infection or reactivation, HCMV is able to infect virtually all organs and tissues, including macrophages, endothelial, epithelial, stromal, neuronal, and smooth muscle cells (Plachter et al., 1996). In vitro, however, full productive infection is supported only in secondary cell strains of fibroblasts and endothelial cells, in fully differentiated myeloid cells, and in primary cultures of hepatocytes and lung epithelial cells (Michelson-Fiske et al., 1975; Sinzger et al., 1995, 1999; Plachter et al., 1996). We demonstrated here that human adrenocortical cell lines are fully permissive for infection and
A
NCI-H295R (HSV-1 F, moi 2)
NCI-H295R (mock infected)
C
200
Cortisol (ng/ml)
Mock
150
HSV-1 F
HSV-1 V-41
10X
100
50
0
4
8
24
48
72
10
20X
17ß-estradiol (ng/ml)
8
6
4
2
B
180
HSV-1 F
0
Cell viability (% of mock infected cells)
160
4
8
24
48
HSV-1 V41
72
140
Aldosterone (pg/mL)
800
120
100
600
80
400
60
40
200
20
0
8
24
48
72
0
4
8
24
48
72
Time (hpi)
Time (hpi)
replication not only by fibroblast-adapted laboratory HCMV strains but also by clinical isolates of HCMV. Moreover, at variance with other epithelial tumor cell lines susceptible to persistent but not lytic HCMV infection (Smith, 1986; Cinatl et al., 1996; Jarvis et al., 1999; Wang and Shenk, 2005), human adrenocortical cells are permissive for lytic HCMV infection, which typically occurs during natural infection. HCMV replication rate in human adrenocortical cell lines was relatively slow and, before death, infected cells showed CPE characterized by cell rounding and shrinking and formation of plaques in cell culture monolayer.
Interestingly, HCMV infection of adrenocortical cells significantly increased the production of cortisol and 17ß- estradiol, while it did not appreciably modulate aldosterone and DHEA-S production. This effect was paralleled by a rapid and marked up-regulation of CYPI IBI and CYP19 expression, the genes responsible for the late steps of cortisol and estrogen production, respectively, whereas CYPI IB2 mRNA increased only at 72 h p.i., and expression of other steroidogenic enzymes was down-regulated. A marked up-regulation of StAR, the key regulator of adrenal steroidogenesis, also occurred in the early phases p.i. These results outline a complex picture of steroidogenesis activation in HCMV-infected cells, involving both the acute control mechanism (induction of StAR expression) and the long-term control mechanism (induction of steroidogenic enzyme expression), but with an atypical pattern, since not all steroidogenic enzymes were up-regulated (Fig. 7B). The expression pattern of transcription factors involved in the regulation of StAR and steroidogenic enzymes expression was also unexpected. In fact, both SF-I and DAX-I, which, respectively, activate and repress StAR expression (Ozisik et al., 2002; Sirianni et al., 2002; lyer and McCabe, 2004; lyer et al., 2006), were down-regulated in HCMV-infected adrenocortical cells, while a repressor of SF-I activity, SHP, was up-regulated. Although this response could not be easily explained in the context of steroidogenesis gene expression and steroid hormone production, it indicated HCMV infection affected also these key regulators of adrenocortical cells at early phases p.i. The effect of HCMV infection on steroidogenesis was probably due to HCMV immediate-early (IE) gene expression, since it was most evident in the early phases of infection and UV- inactivated viral particles did not affect hormone production. We are currently examining if HCMV IEI-72 and IE2-86 can modulate adrenal steroidogenesis. These viral proteins can transactivate many cellular genes and interact with different transcription factors, including Spl and AP-I, which are involved in the regulation of several steroidogenesis factors (Luu and Flores, 1997). In this regard, microarray analysis of gene expression in NCI-H295R adrenocortical cells showed that expression of many genes encoding transcription factors significantly changed in the first hours following HCMV infection. Interestingly, the mechanism of induction of steroidogenesis and steroidogenic gene expression following HCMV infection seems to be different from the mechanism of HCMV activation of the innate immune response. In fact, in fibroblasts and endothelial cells in vitro, HCMV infection induces cellular transcription pathways that activate the production of alpha and beta interferons, cytokines, and proinflammatory chemokines that ultimately lead to the limitation of viral replication and spread (Boehme and Compton, 2006). At variance with induction of steroidogenesis by HCMV, activation of the innate immune response seems to be mediated by the interaction of viral glycoproteins with host cell receptors during the entry phase, while expression of HCMV IE2-86 blocks virus-induced interferon and chemokine expression (Taylor and Bresnahan, 2005, 2006).
We tested the hypothesis that glucocorticoids and other steroid hormones produced by human adrenocortical cells could facilitate HCMV replication, as demonstrated in
fibroblasts and macrophages (Tanaka et al., 1984; Koment, 1989; Lee et al., 1999). But, when adrenocortical cells were treated with hydrocortisone or the general inhibitor of steroidogenesis L-aminoglutethimide, no marked effects were observed on HCMV replication, even though treatment with hydrocortisone seemed to facilitate virus release from cells, as revealed by peaks of increased viral titer in culture medium. These results, however, do not exclude a role for steroid hormones in HCMV replication, since low intracellular hormone levels, such as those produced by adrenocortical cells even in the presence of L-aminoglutethimide, might be sufficient to induce viral replication. On the other hand, HCMV infection of NCI-H295R cells not only increased cortisol and other steroid hormone production, but also led to a significant up-regulation of the glucocorticoid receptor GRa, as shown by microarray experiments, so infected cells could be more sensitive to low levels of glucocorticoids. The mechanism by which glucocorticoids enhance HCMV replication in vivo is still unclear, since a glucocorticoid responsive element has not been identified in HCMV genome. However, the glucocorticoid/GR complex can also modulate transcription by acting as co-activator of cellular transcription factors, such as Spl, NF-KB, AP-I, and CREB/ATF. These transcription factors have their cognate binding sites in the major immediate-early (MIE) HCMV enhancer/promoter (Meier and Stinski, 2006), so, by binding to these transcription factors, glucocorticoids could indirectly enhance viral replication.
Finally, we would like to emphasize that the effect of HCMV infection on adrenal steroidogenesis was HCMV specific, since it was not observed in HSV-I-infected cells, notwithstanding adrenocortical cells appeared to be more susceptible to productive HSV infection than to HCMV infection. Studies in the literature demonstrated HSV-I and HSV-2 efficiently infect and replicate in the adrenal gland after inoculation in mice through different routes (Barzon et al., 2004). In vivo, HSV infection mainly involves the zona fasciculata and induces apoptosis. After intracerebral administration but not after systemic inoculation, HSV-I induces a marked activation of the HPA axis, thus, indicating that the effects on the HPA axis are mediated centrally and not by systemic mechanisms (Barzon et al., 2004). On the other hand, we have recently demonstrated that both wild-type adenovirus and non-replicating adenoviral vectors efficiently infect human adrenocortical cells in vitro, leading to a marked and dose-dependent increase of cortisol and other steroid hormone production and consistently modulated expression of StAR and other steroidogenic enzymes. In this case, induction of steroidogenesis, which was already apparent at 6 h p.i., probably represented a response of the adrenocortical cell to adenovirus entry, but not to viral gene expression, since it was observed with both replicating and non-replicating vectors (Matkovic et al., 2009).
In conclusion, this study demonstrates for the first time that HCMV infects and replicates in human adrenocortical cells, leading to CPE and cell death. Most importantly, HCMV infection of adrenocortical cells leads to a rapid and marked up-regulation of StAR and steroidogenic enzyme expression and to increased cortisol and other steroid hormone production. This early glucocorticoid response to HCMV infection was a direct effect of HCMV infection of adrenocortical cells and not just an indirect effect due to hypothalamus and pituitary stimulation, as suggested in the murine in vivo model (Shanley, 1987; Reddehase et al., 1988; Price et al., 1996; Ruzek et al., 1997, 1999; Silverman et al., 2004). An acute glucocorticoid response is typically triggered by infections and is considered to be critical to host defense against pathogens, although, in the case of HCMV infection, it might also enhance viral replication and reactivation from latency.
EFFECTS OF HCMV INFECTION
Acknowledgments
This work was supported by grant no. RSF 271/07 from Veneto Region to Giorgio Palù and by grant no. 60A06-9578/07 from University of Padova to Luisa Barzon.
Literature Cited
Abate DA, Watanabe S, Mocarski ES. 2004. Major human cytomegalovirus structural protein pp65 (ppUL83) prevents interferon response factor 3 activation in the interferon response. J Virol 78:10995-11006.
Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ. 1997. Gapped BLAST and PSI-BLAST: A new generation of protein database search programs. Nucleic Acids Res 25:3389-3402.
Apweiler R, Bairoch A, Wu CH, Barker WC, Boeckmann B, Ferro S, Gasteiger E, Huang H, Lopez R, Magrane M, Martin MJ, Natale DA, O’Donovan C, Redaschi N, Yeh LS. 2004. UniProt: The universal protein knowledgebase. Nucleic Acids Res 32:D115-119.
Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, Harris MA, Hill DP, Issel-Tarver L, Kasarskis A, Lewis S, Matese JC, Richardson JE, Ringwald M, Rubin GM, Sherlock G. 2000. Gene ontology: Tool for the unification of biology. The Gene Ontology Consortium. Nat Genet 25:25-29.
Barzon L, Boscaro M, Palù G. 2004. Endocrine aspects of cancer gene therapy. Endocr Rev 25:1-44.
Barzon L, Trevisan M, Masi G, Pacenti M, Sinigaglia A, Macchi V, Porzionato A, De Caro R, Favia G, lacobone M, Palù G. 2008. Detection of polyomaviruses and herpesviruses in human adrenal tumors. Oncogene 27:857-864.
Boehme KW, Compton T. 2006. Virus entry and activation of innate immunity. In: Reddehase MJ, editor. Cytomegaloviruses. Molecular biology and immunology. UK: Caister Academic Press. pp 111-130.
Boehme KW, Singh J, Perry ST, Compton T. 2004. Human cytomegalovirus elicits a coordinated cellular antiviral response via envelope glycoprotein B. J Virol 78:1202-121 1. Browne EP, Wing B, Coleman D, Shenk T. 2001. Altered cellular mRNA levels in human cytomegalovirus-infected fibroblasts: Viral block to the accumulation of antiviral mRNAs. J Virol 75:12319-12330.
Chan G, Bivins-Smith ER, Smith MS, Smith PM, Yurochko AD. 2008. Transcriptome analysis reveals human cytomegalovirus reprograms monocyte differentiation toward an MI macrophage. J Immunol 181:698-711.
Cinatl J, Jr., Vogel JU, Cinatl J, Weber B, Rabenau H, Novak M, Kornhuber B, Doerr HW. 1996. Long-term productive human cytomegalovirus infection of a human neuroblastoma cell line. Int J Cancer 65:90-96.
Fontana P, Cestaro A, Velasco R, Formentin E, Toppo S. 2009. Rapid annotation of anonymous sequences from genome projects using semantic similarities and a weighting scheme in gene ontology. PLOS ONE 4:e4619.
Hoshino Y, Nagata Y, Gatanaga H, Hosono O, Morimoto C, Tachikawa N, Nomura K, Wakabayashi T, Oka S, Nakamura T, Iwamoto A. 1997. Cytomegalovirus (CMV) retinitis and CMV antigenemia as a clue to impaired adrenocortical function in patients with AIDS. AIDS 11:1719-1724.
Iyer AK, McCabe ER. 2004. Molecular mechanisms of DAX1 action. Mol Genet Metab 83:60-73.
Iyer AK, Zhang YH, McCabe ER. 2006. Dosage-sensitive sex reversal adrenal hypoplasia congenita critical region on the X chromosome, gene I (DAXI) (NROBI) and small heterodimer partner (SHP) (NROB2) form homodimers individually, as well as DAXI-SHP heterodimers. Mol Endocrinol 20:2326-2342.
Jarvis MA, Wang CE, Meyers HL, Smith PP, Corless CL, Henderson GJ, Vieira J, Britt WJ, Nelson JA. 1999. Human cytomegalovirus infection of caco-2 cells occurs at the basolateral membrane and is differentiation state dependent. J Virol 73:4552-4560.
Koment RW. 1989. Restriction to human cytomegalovirus replication in vitro removed by physiological levels of cortisol. J Med Virol 27:44-47.
Lee CH, Lee GC, Chan YJ, Chiou CJ, Ahn JH, Hayward GS. 1999. Factors affecting human cytomegalovirus gene expression in human monocyte cell lines. Mol Cells 9:37-44.
Luu P, Flores O. 1997. Binding of SPI to the immediate-early protein-responsive element of the human cytomegalovirus DNA polymerase promoter. J Virol 71:6683-6691.
Matkovic U, Pacenti M, Trevisan M, Palù G, Barzon L. 2009. Investigation on human adrenocortical cell response to adenovirus and adenoviral vector infection. J Cell Physiol 220:45-57.
Meier JL, Stinski MF. 2006. Major immediate-early enhancer and its gene products. In: Reddehase MJ, editor. Cytomegaloviruses. Molecular biology and immunology. UK: Caister Academic Press. pp 151-166.
Mengoli C, Cusinato R, Biasolo MA, Cesaro S, Parolin C, Palù G. 2004. Assessment of CMV load in solid organ transplant recipients by pp65 antigenemia and real-time quantitative DNA PCR assay: Correlation with pp67 RNA detection. J Med Virol 74: 78-84.
Michelson-Fiske S, Arnoult J, Febvre H. 1975. Cytomegalovirus infection of human lung epithelial cells in vitro. Intervirology 5:354-363.
Ozisik G, Achermann JC, Jameson JL. 2002. The role of SFI in adrenal and reproductive function: Insight from naturally occurring mutations in humans. Mol Genet Metab 76: 85-91.
Plachter B, Sinzger C, Jahn G. 1996. Cell types involved in replication and distribution of human cytomegalovirus. Adv Virus Res 46:195-261.
Price P, Olver SD, Silich M, Nador TZ, Yerkovich S, Wilson SG. 1996. Adrenalitis and the adrenocortical response of resistant and susceptible mice to acute murine cytomegalovirus infection. Eur J Clin Invest 26:811-819.
Pulakhandam U, Dincsoy HP. 1990. Cytomegaloviral adrenalitis and adrenal insufficiency in AIDS. Am J Clin Pathol 93:651-656.
Rainey WE, Saner K, Schimmer BP. 2004. Adrenocortical cell lines. Mol Cell Endocrinol 228:23-38.
Razzaq F, Dunbar EM, Bonington A. 2002. The development of cytomegalovirus-induced adrenal failure in a patient with AIDS while receiving corticosteroid therapy. HIV Med 3:212-214.
Reddehase M, Jonjic S, Weiland F, Mutter W, Koszinowski UH. 1988. Adoptive immunotherapy of murine cytomegalovirus adrenalitis in the immunocompromised host: CD4-helper-independent antiviral function of CD8-positive memory T lymphocytes derived from latently infected donors. J Virol 62:1061-1065.
Ruzek MC, Miller AH, Opal SM, Pearce BD, Biron CA. 1997. Characterization of early cytokine responses and an interleukin (IL)-6-dependent pathway of endogenous glucocorticoid induction during murine cytomegalovirus infection. J Exp Med 185:1185- 1192.
Ruzek MC, Pearce BD, Miller AH, Biron CA. 1999. Endogenous glucocorticoids protect against cytokine-mediated lethality during viral infection. J Immunol 162:3527-3533.
Shanley JD. 1987. Modification of acute murine cytomegalovirus adrenal gland infection by adoptive spleen cell transfer. J Virol 61:23-28.
Shanley JD, Lutwick LI, Donta ST. 1979. Replication of murine cytomegalovirus in murine Y-I cells. J Med Virol 4:261-268.
Silverman MN, Miller AH, Biron CA, Pearce BD. 2004. Characterization of an interleukin-6- and adrenocorticotropin-dependent, immune-to-adrenal pathway during viral infection. Endocrinology 145:3580-3589.
Simmen KA, Singh J, Luukkonen BG, Lopper M, Bittner A, Miller NE, Jackson MR, Compton T, Früh K. 2001. Global modulation of cellular transcription by human cytomegalovirus is initiated by viral glycoprotein B. Proc Natl Acad Sci USA 98:7140-7145.
Sinzger C, Grefte A, Plachter B, Gouw AS, The TH, Jahn G. 1995. Fibroblasts, epithelial cells, endothelial cells and smooth muscle cells are major targets of human cytomegalovirus infection in lung and gastrointestinal tissues. J Gen Virol 76:741-750.
Sinzger C, Bissinger AL, Viebahn R, Oettle H, Radke C, Schmidt CA, Jahn G. 1999. Hepatocytes are permissive for human cytomegalovirus infection in human liver cell culture and in vivo. J Infect Dis 180:976-986.
Sirianni R, Seely JB, Attia G, Stocco DM, Carr BR, Pezzi V, Rainey WE. 2002. Liver receptor homologue-I is expressed in human steroidogenic tissues and activates transcription of genes encoding steroidogenic enzymes. J Endocrinol 174:R13-R17.
Slobedman B, Stern JL, Cunningham AL, Abendroth A, Abate DA, Mocarski ES. 2004. Impact of human cytomegalovirus latent infection on myeloid progenitor cell gene expression. J Virol 78:4054-4062.
Smith JD. 1986. Human cytomegalovirus: Demonstration of permissive epithelial cells and nonpermissive fibroblastic cells in a survey of human cell lines. J Virol 60:583-588.
Tanaka J, Ogura T, Kamiya S, Sato H, Yoshie T, Ogura H, Hatano M. 1984. Enhanced replication of human cytomegalovirus in human fibroblasts treated with dexamethasone. J Gen Virol 65:1759-1767.
Taylor RT, Bresnahan WA. 2005. Human cytomegalovirus immediate-early 2 gene expression blocks virus-induced beta interferon production. J Virol 79:3873-3877.
Taylor RT, Bresnahan WA. 2006. Human cytomegalovirus immediate-early 2 protein IE86 blocks virus-induced chemokine expression. J Virol 80:920-928.
Wang D, Shenk T. 2005. Human cytomegalovirus ULI31 open reading frame is required for epithelial cell tropism. J Virol 79:10330-10338.
Zhu H, CongJP, Shenk T. 1997. Use of differential display analysis to assess the effect of human cytomegalovirus infection on the accumulation of cellular RNAs: Induction of interferon- responsive RNAs. Proc Natl Acad Sci USA 94:13985-13990.
Zhu H, Cong JP, Mamtora G, Gingeras T, Shenk T. 1998. Cellular gene expression altered by human cytomegalovirus: Global monitoring with oligonucleotide arrays. Proc Natl Acad Sci USA 95:14470-14475.