Investigation on Human Adrenocortical Cell Response to Adenovirus and Adenoviral Vector Infection
URSKA MATKOVIC, MONIA PACENTI, MARTA TREVISAN, GIORGIO PALÙ,* AND LUISA BARZON*
Department of Histology, Microbiology, and Medical Biotechnologies, University of Padova, Padova, Italy
After systemic administration, adenoviral vectors (AdVs) are sequestered in the liver and adrenal glands. Adenoviral vector transduction has been shown to cause cytopathic effects on human hepatocytes and to induce an inflammatory response, whereas the effect of AdVs on human adrenocortical cells has never been investigated. In this study, human adrenocortical carcinoma cell lines and primary cell cultures were used to assess the effects of wild-type adenovirus (Ad5WT) and EI/E3-deleted AdVs on cell proliferation and steroidogenesis. Ad5WT could efficiently replicate in adrenocortical cells, leading to S phase induction, followed by cell death, whereas high titer AdVs transduction had only mild effects on adrenocortical cell proliferation, with accumulation of cells in G2/M. Both AdVs and Ad5WT induced expression of inflammatory cytokines, such as interleukin-6 and tumor necrosis factor-a, but, most importantly, they led to a marked and dose-dependent increase of cortisol and other steroid hormone production and consistently modulated expression of key steroidogenic enzymes and regulators of steroidogenesis. This effect, which was already apparent at 6 h post-infection, probably represented a response to adenoviral entry and/or early phases of infection. The result of this study contribute to the understanding of host response to adenoviral vector administration.
J. Cell. Physiol. 220: 45-57, 2009. @ 2009 Wiley-Liss, Inc.
A number of gene therapy clinical trials using recombinant adenoviral vectors (AdVs) have established a promising therapeutic potential of these vectors, which can be easily produced at high titer and can efficiently transduce a variety of dividing and non-dividing cells in vivo (Thomas et al., 2003). But, a major obstacle to the clinical application of AdVs is the innate immune response elicited in the host against the vector, which limits vector safety and efficacy (Thomas et al., 2003). After systemic administration, adenoviral infection induces a rapid and marked innate and adaptive immune response, which depends on virus capsid interaction with host cells (Liu and Muruve, 2003). Studies in animal models and in humans have shown that the innate immune response to adenovirus and AdV infection is characterized by complement activation, secretion of proinflammatory cytokines (such as tumor necrosis factor [TNF]-&, interleukin-6 [IL-6], IL-1, IL-8, interferon-y [IFN-y], and type | IFNs a and B) and chemokines (such as macrophage inflammatory proteins 1 and 2) by dendritic cells and macrophages, and activation of natural killer cells (Liu and Muruve, 2003). This innate immune response, which is responsible for vector AdV toxicity and is characterized by vascular and tissue damage and even death (Mickelson, 2000), has been mainly ascribed to sequestration of adenoviral particles into the liver with infection of Kupffer cells and hepatocytes (Shayakhmetov et al., 2004, 2005).
Besides liver targeting, adenoviruses have a natural tropism for the adrenal gland, as shown after systemic administration in mice and in non-human primates (reviewed by Barzon et al., 2004). In particular, efficient transgene expression in the adrenal glands has been demonstrated after intravenous AdV injection in mice (Groot-Wassink et al., 2002). Within the adrenal gland, zona fasciculata of the adrenal cortex, where cortisol and corticosterone are synthesized, is most susceptible to adenoviral infection (Wang et al., 2003).
While liver toxicity due to AdV infection has been extensively investigated, there is scarce data about the effects of adenoviruses and AdVs on the adrenal gland (Barzon et al., 2004). Severe morphological and ultrastructural changes were
described in the adrenal glands of calves and rodents after experimental adenovirus infection (Hoenig et al., 1974; Margolis et al., 1974; Cutlip and Mcclurkin, 1975; Strauber and Card, 1978), as well as in infants who died after disseminated adenoviral infection (Medvedev and Shastina, 1978). More recently, Alesci et al. (2002) showed that AdV transduction of bovine adrenocortical cells in primary cultures increased cell proliferation and led to nuclear fragmentation and mitocondrial alterations. Moreover, AdV infection increased basal steroid hormone production, but decreased steroid hormone release after adrenocorticotropin (ACTH) stimulation (Alesci et al., 2002).
Besides being a target of AdV toxicity, the adrenal gland could be involved in the innate immune response to systemic adenoviral infection. It is well known that activation of the hypothalamic pituitary adrenal (HPA) function, which results in increased cortisol production, has an important role in host response and is critical for host survival during acute and chronic stressful events such as sepsis and viral infections
Giorgio Palù and Luisa Barzon contributed equally to this work. Contract grant sponsor: Ministero dell’istruzione dell’Università e della Ricerca (MIUR);
Contract grant number: 2004063894_003. Contract grant sponsor: Veneto Region; Contract grant number: RSF 271/07.
*Correspondence to: Giorgio Palù and Luisa Barzon, Department of Histology, Microbiology, and Medical Biotechnologies, University of Padova; Via A. Gabelli 63; 1-35128 Padova, Italy. E-mail: giorgio.palu@unipd.it; luisa.barzon@unipd.it
Received 10 November 2008; Accepted 9 January 2009
Published online in Wiley InterScience (www.interscience.wiley.com.), 6 February 2009. DOI: 10.1002/jcp.21727
(Arafah, 2006). During critical illness, in addition to hypothalamic hormones, inflammatory cytokines such as IL-I, IL-6, and TNF-a, acting at both pituitary and adrenocortical levels, are capable of stimulating cortisol production, which at high concentrations has antiinflammatory effects on the immune system and can modulate hepatic acute phase response (Arafah, 2006).
The recent demonstration that human adrenocortical cells express Toll-like receptors (TLRs) further supports the role of the adrenal gland in the early defense against infections (Bornstein et al., 2004b; Zacharowski et al., 2006; Tran et al., 2007). Among the TLR family members identified in humans, TLR9, which recognizes unmethylated CpG motifs present in bacterial as well as in viral dsDNA, has been demonstrated to mediate the innate immune response to adenovirus and AdV infection in dendritic cells (Zhu et al., 2007) and macrophages (Cerullo et al., 2007), as well as in non-immune cells, such as respiratory epithelial cells (Hartmann et al., 2007) and could be involved in adrenal response to viral infection. In fact, challenge with CpG synthetic oligonucleotides (CpG-ODN) resulted in a significant increase of plasma corticosterone and inflammatory cytokine levels in wild-type but not in TLR-9 deficient mice (Tran et al., 2007), suggesting this receptor play a role in adrenal stress response in conditions in which microbial DNA is present.
To our knowledge, the effect of AdVs on human adrenocortical cells, and in particular on TLR pathways, cytokine release, and steroid hormone production, has never been investigated. In this study, we analyzed the direct effects of wild-type adenovirus serotype 5 (Ad5WT) and first-generation AdVs on human adrenocortical cells. The study was performed in vitro to avoid the indirect effects of activation of the HPA axis and immune-inflammatory cells that typically occur in vivo. In addition to the examination of basic cell parameters such as cell proliferation, cell cycle profile, and cell death, particular attention was focused on adenovirus-mediated modulation of adrenal steroidogenesis and immune response induced by TLR-9. We show here that both AdVs and Ad5WT infection of human adrenocortical cells induced expression of inflammatory cytokines, but, most importantly, they led to a marked increase of cortisol and other steroid hormone production and consistently modulated expression of key steroidogenic enzymes and regulators of steroidogenesis.
Materials and Methods Cell lines
SW-13 and NCI-H295R cell lines, both derived from human adrenocortical carcinomas, were obtained from the American Type Culture Collection (ATCC, Rockville, MD). SW-13 cells were cultured in Leibovitz’s L-15 medium supplemented with 10% fetal calf serum (FCS) and penicillin/streptomycin (P/S) 1% at 37℃ in humidified incubator without CO2. The NCI-H295R cell line was grown in RPMI medium supplemented with 2% FCS, 1% insulin, transferrin, selenium (ITS) and P/S 1% at 37°℃ in humidified incubator with 5% CO2. HEK 293 cells (ATCC) were grown in DMEM supplemented with 10% FCS and P/S 1% at 37℃ and 5% CO2. Media and supplements were purchased from Invitrogen S.r.l. (Milan, Italy).
Primary human adrenocortical cell cultures
Primary cultures of human adrenocortical carcinoma were prepared from five 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 protease 1 mg/ml, DNAse 67 mg/ml, Amphotericin B 1%, and P/S 1%. 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%, ITS 1%, and FCS 10%. For the experiments, cells were seeded on 24-well plates at a density of 2 × 105 cells per well. Cells were washed gently with phosphate buffer saline and growth medium was replaced every 24 h. After 1-2 days in culture, about 50% confluent cells were infected with Ad5 and AdEGFP, at a multiplicity of infection (MOI) of about 100 plaque forming units (pfu) per cell, to assess the susceptibility for adenoviral infection by fluorescent microscopy and to measure steroid production in growth medium as below described.
Ad5WT and AdV production and titration
Ad5WT was obtained from ATCC. Replication-incompetent AdVs based on the Ad5 genome and lacking the El and E3 regions were constructed by homologous recombination in E. coli using AdEasy vector system (Qbiogene, Carlsbad, CA). In these vectors, human cytomegalovirus promoter was used to drive expression of the transgenes encoding herpes simplex virus thymidine kinase (AdHSV-TK) and green fluorescent protein (AdEGFP). The Adnull vector (Qbiogene), which is identical to the above AdVs but it contains no transgene, was used as a control of transgene toxicity. AdVs were propagated in El-complementing HEK 293 cells, purified by cesium chloride density centrifugation, and titrated by TCID50 cpe endpoint assay according to the AdEasy production protocol. Viral vector stocks were stored at 5 x 109 pfu/ml concentration in 10% glycerol at -80℃ until use. Viral infections are given as MOI, expressed as pfu per cell. In all cases, the uninfected mock controls were treated under the same conditions as for infection but for the presence of the virus.
Adenovirus infectivity test
AdEGFP and Ad5WT were used to test adenovirus infectivity in both adrenocortical cell lines. NCI-H295R and SW-13 cells were plated on 6-well plates at a density of 5 x 105 and 3 x 105 cells per well, respectively. After 24 h, cells were infected with increasing MOIs of AdEGFP, Ad5WT, or mock infected; at 48 h pi cells were harvested, washed in PBS, and the percentage of EGFP-positive cells was analyzed by flow cytometry in the case of AdEGFP infection or by immuno-fluorescence staining with a mouse monoclonal antibody for adenovirus hexon structural protein included the VRK Bartels Viral Respiratory Screening and Identification Kit (Trinity Biotech plc, Wicklow, Ireland) in the case of infection with Ad5WT.
Analysis of adenovirus replication kinetics
NCI-H295R and SW-13 cells were seeded as above reported and infected with Ad5WT at MOI 5 or mock infected. Media were harvested at described time points pi and used to determine viral titer by standard TCID50 plaque assay. In addition, viral replication in NCI-H295R and SW-13 cells was evaluated by immuno- fluorescence staining against adenovirus hexon protein as above reported.
Cell cycle analysis
SW-13 and NCI-H295R cells were disseminated at a density of 2 x 105 and I x 106 cells per 25-cm2 flask, respectively. At this density, cells reached about 70% confluence. To synchronize the cell population, growth medium was changed with serum-free medium for 24 h before infection. Cells were mock infected and infected with AdVs at different MOIs and harvested in time-course experiments. Cells were washed in PBS and fixed for 30 min in cold 70% EtOH at -20℃. After washing in phosphate citrate buffer and in PBS, fixed cells were treated with 0.1 mg/ml RNase in PBS at 37℃ for 30 min. Thereafter, cells were stained with 0.1 mg/ml propidium iodide (Sigma-Aldrich Srl, Milan, Italy) in PBS at room temperature for 30 min and analyzed by flow cytometry using a Becton-Dickinson FACScan (Becton-Dickinson, Franklin Lakes, NJ).
Apoptosis detection
SW-13 and NCI-H295R cells were infected with Ad5WT and AdVs at different MOIs and harvested at 24, 48, and 72 h pi for annexin V staining using the Vybrant Apoptosis Assay Kit (Molecular Probes, Invitrogen S.r.l.). Percentages of annexin V-positive and propidium iodide-positive cells were determined on a Becton-Dickinson FACScan.
Cell survival analysis
SW-13 and NCI-H295R cells were plated on 96-well plates at 1 x 103 and 5 x 103 cells/well, respectively, to reach about 70% confluence, and, after 24 h, infected with Ad5 and AdVs at MOIs ranging from 2 to 500. Cell survival was assayed in time-course experiments by MTT assay (Sigma-Aldrich) and by BrdU Cell Proliferation Assay (Calbiochem, Darmstadt, Germany).
Steroid hormone measurements
NCI-H295R cells were grown until approximately 80% confluence on 24-well plates. After 1-day growth in serum-free medium, the cells were infected with AdVs and Ad5WT and mock-infected. Cortisol, aldosterone, and 17ß-estradiol were measured in cell culture supernatant at 6, 12, 24, 48, 72, and 96 h pi by using EIA kits (Cayman, Ann Arbor, MI).
Quantitative real-time RT-PCR analysis of gene expression
Expression of mRNA encoding steroidogenic enzymes (CYP21, CYP19, CYPIIBI, CYPI IB2), regulators of steroidogenesis (StAR, SF-I, DAX), cytokines (IL-6, TNF-a, IFN-a, IFN-B), IFN receptors, TLR-9, and IRF7 was analyzed by quantitative real-time RT-PCR in NCI-H295R cells after infection with Ad5WT and AdVs and challenge with CpG and LPS. Briefly, total RNA was purified from cells and tissue samples using the RNeasy kit (Qiagen S.p.A., Milan, Italy). Random primed cDNA was synthesized from total RNA using MuLV reverse transcriptase (Applied Biosystems, Foster City, CA) and used for quantitative real-time RT-PCR, which was performed using SYBR Green PCR Master Mix or TaqMan Master Mix (Applied Biosystems) on an ABI PRISM 7900 HT Sequence Detection System (Applied Biosystems) as reported (Barzon et al., 2008). Absolute quantification was performed against a standard curve obtained by amplification of correspondent DNA sequences subcloned into the pCR2.1 vector (Invitrogen Life Technologies S.p.A, San Giuliano Milanese, Italy). mRNA copy number was calculated automatically by 7900 ABI PRISM SDS software against the standard curve. mRNA values are reported as copies/µg total RNA, after standardization against the housekeeping gene GAPDH, which was also absolutely quantified by real-time RT-PCR. Oligonucleotide primer and probe sequences and annealing temperatures are reported in Table 1.
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
Human adrenocortical cells are susceptible to adenoviral infection
The SW-13 and NCI-H295R cells were employed to investigate the effects of adenoviral infection on human adrenocortical cells. SW-13 cells are derived from a poorly differentiated non-functioning carcinoma carrying several chromosomal abnormalities and display fully malignant phenotype, with a rapid growth rate (Leibovitz et al. 1973). NCI-H295R cells derive from a functioning adrenocortical carcinoma, have a slow growth rate and, cytogenetically, are highly aneuploid (Gazdar et al. 1990). NCI-H295R cells express genes encoding all adrenal steroidogenic enzymes and retain the ability to produce aldosterone, cortisol and C19 steroids and to respond to most physiologic stimuli, such as angiotensin II, potassium ions, and cAMP analogues (Rainey et al., 1994, 2004).
To test the susceptibility of human adrenocortical cells in vitro to adenovirus infection and to compare transduction efficiency of first generation EI/E3-deleted AdVs with Ad5WT, SW-13 and NCI-H295R cells were transduced with Ad5WT and with AdEGFP at different MOIs and were analyzed by fluorescent microscopy and flow cytometry for Ad5 hexon protein and EGFP expression, respectively. Efficiency of AdEGFP and Ad5WT transduction was high in both cell lines and was increasing with higher MOIs (Table 2). In addition to adrenocortical carcinoma cell lines, Ad5 and AdEGFP efficiently infected primary cell cultures established from adrenocortical adenomas and carcinomas, as shown by immuno-staining against Ad5 hexon protein and by fluorescence microscopy (Fig. IA). These results confirm the high susceptibility of human adrenocortical cells to adenoviral infection. Moreover, efficiency of AdV transduction was at the similar level as efficiency of Ad5WT infection, indicating the internalization mechanism of replication-incompetent AdVs remains intact.
To analyze adenovirus replication kinetics in adrenocortical cells, Ad5WT production was measured in supernatant of infected cells over different time points pi. Ad5WT replicated efficiently in both SW13 and NCI-H295R cells, as compared with control HEK 293 cells, but displayed different replication kinetics in the two cells lines. In fact, the maximal viral titer was reached at day 3 pi (1.3 x 109 pfu/ml) and at day 7 pi (1.7 x 108 pfu/ml) in SW-13 and NCI-H295R cells, respectively
| Gene | Forward primer (5' -> 3') | Reverse primer (5' -> 3') | TaqMan probe (5'-> 3') | Annealing temperature (C) |
|---|---|---|---|---|
| StAR | CCACCCCTAGCACGTGGA | TCCTGGTCACTGTAGAGAGTCTCTTC | 60 | |
| SF1 | GGAGTTTGTCTGCCTCAAGTTCA | CGTTCTTTCACCAGGATGTGGTT | 60 | |
| DAX1 | CCAAGGAGTACGCCTACCTCAA | ACTGGAGTCCCTGAATGTACTTCC | 60 | |
| CY11A | TCCAGAAGTATGGCCCGATT | CATCTTCAGGGTCGATGACATAAA | 60 | |
| CYP2 1 | TCAGGTTCTTCCCCAATCCA | TCCACGATGTGATCCCTCTTC | 60 | |
| CYP11B1 | GGCAGAGGCAGAGATGCTG | TCTTGGGTTAGTGTCTCCACCTG | TGCTGCACCATGTGCTGAAACACCT | 58 |
| CYP11B2 | GGCAGAGGCAGAGATGCTG | CTTGAGTTAGTGTCTCCAGGA | CTGCACCACGTGCTGAAGCACT | 68 |
| CYP19 | TCACTGGCCTTTTTCTCTTGGT | GGGTCCAATTCCCATGCA | 60 | |
| TLR-9 | TGTGAAGCATCCTTCCCTGT | GAGAGACAGCGGGTGCAG | 60 | |
| IRF-7 | AGCTGTGCTGGCGAGAAG | TGTGTGTGCCAGGAATGG | 60 | |
| IFN-& | AGAATCTCTCHTTYCTCCTG | TTCTGCTCTGACAACCTCC | 58 | |
| IFN-B | TCTAGCACTGGCTGGAATGAG | GTTTCGGAGGTAACCTGTAAG | 58 | |
| IFNAR-1 | ATTTACACCATTTCGCAAAGC | CACTATTGCCTTATCTTCAGCTTCTA | 59 | |
| IFNAR-2 | AACGTTGTTCAGTTGCTCACA | TCTCAAACTCTGGTGGTTCAAA | 59 | |
| IL-6 | GGGAAGGTGAAGGTCGG | TGGACTCCACGACGTACTCAG | 60 | |
| TNF-a | CCCAGGGACCTCTCTCTAATC | ATGGGCTACAGGCTTGTCACT | 60 | |
| GAPDH | GAAGGTGAAGGTCGGAGTC | GAAGATGGTGATGGGATTTC | CAAGCTTCCCGTTCTCAGCC | 60 |
| Cell line/AdV | Mean percentage (±SD) of Ad-positive cellsa | |||||||
|---|---|---|---|---|---|---|---|---|
| MOI 5 | MOI 10 | MOI 15 | MOI 25 | MOI 50 | MOI 100 | MOI 300 | MOI 500 | |
| SW-13/AdEGFP | 60.2 ±7.3 | 70.1±8.4 | 75.5 ±7.2 | 76.6 ± 10.1 | 82.0 ±9.6 | 91.3 ±8.6 | 96.0 ±5.3 | 96.6± 4.3 |
| SW-13/Ad5WT | 58.1±9.2 | 73.0 ±9.1 | 76.8± 8.4 | 77.9 ±9.3 | 80.5 ±10.2 | 93.2 ±7.2 | 95.3 ±4.7 | 95.5 ± 3.2 |
| NCI-H295R/AdEGFP | 59.5±6.8 | 64.5 ±7.2 | 72.5 ±5.0 | 75.6 ± 6.3 | 79.8 ±8.2 | 92.5 ±6.2 | 96.1 ±4.2 | 97.1±2.4 |
| NCI-H295R/Ad5WT | 55.3 ± 10.2 | 67.2 ±8.4 | 70.8 ±7.0 | 74.8 ±8.9 | 76.6 ± 10.3 | 90.8 ±7.2 | 92.4 ±5.2 | 94.4 ±5.1 |
*MOIs are expressed as pfu/cell.
ªResults represent mean percentages ±SD of data from experiments performed three times in duplicate.
(Fig. 1B). This result was confirmed by immunofluorescence staining for adenovirus structural protein, which showed an increasing amount of Ad5WT-positive cells with time, indicating production and spread of the virus (Fig. IB). Infectivity tests and replication kinetics experiments enabled to design time courses of subsequent experiments, considering the time of maximal virus production and growth rate of the two cell lines.
Cytopathic effects of adenoviral infection on human adrenocortical cells
Infection with AdVs deleted in E1 and E3 regions has been reported to provoke growth retardation of different cell types, microscopically seen as decreased cell confluence with rounded nonadherent cells (Brand et al., 1999). We examined whether human adrenocortical cells as well undergo typical morphological changes upon AdV transduction. In this experiments, AdVs carrying different transgenes (AdEGFP and AdHSV-TK) or without transgene (Adnull) were used to exclude the possibility that AdV-induced phenotypic changes of infected cells was specific to the transgene. In NCI-H295R cells, altered monolayer integrity with decrease in cell density and increase in cell size was observed after infection with all three AdVs, indicating that this effect was not restricted to the transgene (Fig. 2). At variance, no marked changes in cellular morphology were observed in SW-13 cells after infection (Fig. 2).
The effects of recombinant AdVs on human adrenocortical cell proliferation were evaluated by both MTT and BrdU assays, which gave similar results. At low MOIs (from 2 to 50) AdV transduction had no effect on SW-13 proliferation compared with uninfected control, whereas a slight decrease of cell viability was demonstrated in NCI-H295R cells in dose- dependent manner. When cells were infected with higher MOls (MOIs 100 and 500), both cell lines showed reduced survival (about 20% reduction) as compared to uninfected control cells (Fig. 3). The effect on adrenocortical cell survival was also analyzed after Ad5WT infection, which led to time- and dose-dependent decrease of cell viability. Time necessary for Ad5WT to kill 50% of cell population at MOI 50 was 3 days in the case of SW-13 cells and 7 days in the case of NCI-H295R cells (Fig. 3). This result was consistent with replication kinetics of Ad5WT in both cell lines.
Cell cycle analysis after AdV infection at low MOI (from 15 to 50) showed no alteration of cell cycle phases, whereas infection at higher MOI (from 100 to 500) increased the G2/M cell fraction; this effect was most prominent at 2 days pi.
Representative results with both NCI-H295R and SW-13 cells are shown in Figure 4a. Cells were sub-confluent at the time of infection and had not yet reached the confluence at the time of harvesting to rule out contact inhibition was the reason for cell cycle arrest. By annexin V/propidium iodide staining, a significant increase of cell death was found only in NCI-H295R cells, but not in SW-13 cells, after infection with both Adnull and AdV carrying transgenes at high MOI (100-500) (Fig. 4b). At
variance with non-replicating AdVs, infection with Ad5WT, even at low MOI (5-25), induced the S phase of cell cycle in both NCI-H295R and SW-13 cells (Fig. 4c), followed by cell death as a result of viral replication. The maximal increase of cell death was observed at 72 h pi (Fig. 4d).
Adenoviral infection induce adrenal steroidogenesis
Adenoviral vector infection has been demonstrated to impair steroidogenesis in bovine adrenocortical cells and to alter their cellular ultrastructure (Alesci et al., 2002), but no data are available on the effect of adenoviral infection on steroidogenesis of human adrenocortical cells. To investigate this issue, NCI-H295R cells (which can produce all adrenal mineralcorticoids, glucocorticoids, androgens, and estrogens) were infected with AdVs and with Ad5WT and steroid hormones were measured in culture medium. As shown in Figure 5a-c, dose-dependent increase of steroid hormone levels was found 24 h pi with AdEGFP. Cortisol and 17ß- estradiol were markedly increased, whereas aldosterone response to AdEGFP infection was less pronounced. Production of steroid hormones was increased at both early (6 h) and late (24 h) time points pi (Fig. 5d-f) and was still evident in later time points pi (72 and 96 h pi) with AdVs, whereas steroid hormone production decreased at 72 h pi in Ad5WT-infected cells, probably due to cell death (data not shown). Both AdEGFP and Adnull vectors, as well as Ad5WT, demonstrated similar effects on the level of secreted hormones in the early phases of infection, whereas incubation with LPS and CpG-ODN did not show marked effects on hormone production (Fig. 5d-f). A three to fivefold increase of cortisol secretion was also demonstrated in the primary cell cultures established from functioning adrenocortical carcinomas after infection with AdEGFP MOI 20.
To investigate the mechanism by which adenovirus infection affects steroidogenesis, expression of key steroidogenic enzymes and regulators of steroidogenesis was analyzed (Fig. 6). Expression of the gene encoding the steroidogenic acute regulatory protein StAR, a global activator of steroidogenesis, was significantly upregulated by both AdVs and Ad5WT. StAR mRNA expression levels were highest at 6 h pi, but still markedly increased at 24 h pi. Transcript levels of DAX-I (dosage-sensitive sex reversal, adrenal hypoplasia critical region, on chromosome X, gene 1), which is a nuclear receptor that acts as a negative regulator of StAR and steroidogenesis, were unaltered at 6 h pi but markedly downregulated at 24 h pi. On the contrary, mRNA expression of SF-I (steroidogenic factor I), an inductor of StAR and of steroidogenesis, showed a threefold increase with respect to control cells already at 6 h pi (Fig. 6c). Notably, consistent with SF-I and StAR induction, mRNA expression of the steroidogenic enzymes CYP21 (21- hydroxylase), CYPIIBI (IIB-hydroxylase), CYPI IB2 (aldosterone synthase), and CYP19 (aromatase) was significantly induced at early time points pi. At variance, LPS and CpG-ODN had no significant effects on expression of genes involved in steroidogenesis.
A
a
b
C
d
B
10
10
10 q
titer yield (VP/ml)
10
8
1
107
10
6
+-HEK293
+- SW13
5
10
-&-NCI-H295R
10
4
0
1 2 3 4 5 6 7 8 9 10 days post-infection
24h
48h
72h
96h
H295R/mock
H295R/Admull
H295R/AdEGFP
·H295R/AdHSV-TK
Scale Bar = 100 um
SW-13/mock
SW-13/Adnull
SW-13/AdEGFP
SW-13/AdHSV-TK
Scale Bar = 100 um
Adenoviral infection modulates cytokine expression in human adrenocortical cells
To elucidate the mechanism of adenovirus-mediated induction of adrenal steroidogenesis, we investigated TLR pathway and cytokine response to adenoviral infection in NCI-H295R cells. Since production of steroid hormones may be regulated by several cytokines, produced either by adrenal cells or by intra- adrenal and circulating macrophages and blood cells (Bornstein et al., 2004a), we wondered whether increased secretion of steroid hormones induced by adenoviral infection was mediated by intraadrenal inflammatory cytokines and type I IFNs, activated through TLR-9 signaling (Benihoud et al., 2007; Zhu et al., 2007). Thus, we studied expression of TLR-9 and its downstream effector IRF7 (interferon regulatory factor 7, which is a master regulator of type I IFN response following viral infection) in NCI-H295R cells after infection with AdVs and Ad5WT (Fig. 7). Levels of TLR-9 mRNA were low in NCI-H295R cells, as also observed by Tran et al. (2007), and did not show any significant modification by either adenovirus infection or by CpG-ODN and LPS challenge. At variance, IRF7 mRNA significantly increased after AdV and Ad5WT infection,
AdEGFP
NCI-H295R
SW-13
120
Cell viability (% of mock infection)
120
100
100
80
80
£
1
60
£
MOI 2
60
40
MOI 10
MOI 50
40
20
MOI 100
20
MOI 500
0
0
1
2
3
4
5
6
7
8
0
1
2
3
4
Ad5WT
NCI-H295R
SW-13
Cell viability (% of mock infection)
120
120
100
100
80
80
60
60
40
-0-MOI 2
40
20
MOI 10
4- MOI 50
20
0
0
1
2
3
4
5
6
7
8
0
1
2
3
4
Time (days pi)
Time (days pi)
but not after CpG-ODN and LPS challenge. We also analyzed mRNA expression of type I IFNs and their receptor, IL-6, and TNF-a after infection with Ad5WT and AdVs at early (6 h) and late (24 h) time points pi. Regarding IFN response, at baseline NCI-H295R cells expressed low levels of both type I IFNs (IFN-a and IFN-B) mRNAs as well as of type I IFN receptor subunits IFNAR-I and IFNAR-2, which did not significantly change after Ad5WT and AdVs infection nor after CpG-ODN and LPS treatment. Expression of IL-6 and TNF-& mRNA was also low in baseline conditions, but it was markedly induced by adenovirus infection (both AdVs and Ad5WT), but not by CpG and LPS challenge. IFN-&, IFN-ß, and TNF-& proteins were measured in cell culture supernatants by ELISA, but their levels were very low and not quantifiable both at baseline and after viral infection.
Discussion
This in vitro study investigated for the first time the effects of replication-defective AdVs and replication-competent
wild-type adenovirus on human adrenocortical cell growth and steroidogenesis. The data presented above show the high susceptibility of human adrenocortical cells to adenoviral transduction. While AdV infection had mild effects on adrenocortical cell morphology and proliferation, both AdVs and wild-type adenovirus led to a marked increase of cortisol production and consistently modulated expression of key steroidogenic enzymes and regulators of steroidogenesis during the early phases of infection. Moreover, like cells of the immune system, adrenocortical cells responded to adenovirus infection with production of inflammatory cytokines. These data suggest the adrenal cortex could be involved in the innate immune response to adenoviral infection and represent the basis for further investigation in in vivo models.
Regarding the analysis of cytopathic effects of adenoviral infection on adrenocortical cells, the results of this study show that Ad5WT induced the S phase, before the newly produced viral particles provoked cell death, in agreement with the literature (Lowe and Ruley, 1993). The S phase is typically
a
mock
Adnull MOI 100
Adnull MOI 500
255
Go/G1
s-Go:
2.5%
255
H GO/G1
s-Go: 3%
255
s-Go: 2%
G1/G:
70%
G1/G: 61%
G1/G : 41%
NCI-H295R Cell number
s-Go
S: 16%
s-Go
S: 18%
Gg/G1
S: 24%
G2/M:
12%
Cell number
S Gz/M
G2/M: 19%
Cell number
G2/M: 31%
s-Go
S Gz/M
S Gz/M
-
0
Fluorescence intensity
1023
0
0
Fluorescence intensity
1023
0
0 Fluorescence intensity
1023
300
Go/G1
GOG1
s-Go: 2%
300
s-Go: 3%
300
Gg/G1
s-Go: 4%
s-Go
G1/G : 69%
G1/G: 63%
S:
12%
s-Go
G1/G : 44%
SW-13
S: 12%
s-Go
S:
10%
Cell number
G2/M: 14%
Cell number
G2/M: 20%
Cell number
G2/M: 36%
S Gz/M
S Gz/M
S Gz/M
0
0
Fluorescence intensity 1023
0
0
Fluorescence intensity 1023
-
0
Fluorescence intensity 1023
b
mock
Adnull MOI 100
Adnull MOI 500
104
9%
17%
104
12%
21%
104
9%
34%
NCI-H295R
103
103
103
102
102
102
Propidium iodide
101
9%
101
9%
101
15%
10°
100
100
100
101
102
103
104
100
101
102
103
104
10º
101
102
103
104
104
10%
12%
104
13%
9%
104
11%
15%
103
103
103
SW-13
102
102
102
101
8%
3%
10%
101
101
10º
10º
100
10º
101
102
103
104
100
101
102
103
104
10º
101
102
103
104
Annexin V
C
SW-13/mock
SW-13/Ad5WT MOI 25
250
Go/G1
250
s-Go
Cell number
Cell number
Go/G1
S Ga/M
s-Go
S GZIM
-
-
Fluorescence intensity 1023
0 Fluorescence intensity 1023
d
SW-13/mock
SW-13/Ad5WT MOI 25
104
5.7%
9%
104
1%
5.8%
103
103
24 h p.i.
102
102
101
7%
101
15%
100
10º
10º
101
102
103
104
100
101
102
103
104
104
0.7%
4%
104
0.6%
10.5%
103
103
48 h p.i.
102
102
101
7.8%
101
13%
100
10º
10º
101
102
103
104
10º
101
102
103
104
104
0.5%
6%
104
1.2%
40%
Propidium iodide
103
103
72 h p.i.
102
102
101
13%
101
18.7%
100
10º
101
102
100
103
104
100
101
102
103
104
Annexin V
a
250
d
CTR-
*
200
AdEGFP
1400
ctr-
*
1200
Adnull
*
Aldosterone [pg/ml]
*
I
Aldosterone [pg/ml]
AdEGFP
*
150
1000
Ad5
800
LPS
CpG
100
600
400
50
200
0
0
b
ctr-
MOI15
MOI50
MOI100 MOI500
6h
24h
25
e
250
*
*
20
200
*
Cortisol [ng/ml]
Cortisol [ng/ml]
15
*
150
T
*
10
*
100
5
50
*
*
*
0
0
I
ctr-
MOI15
MOI50
MOI100
MOI500
6h
24h
C
25
f
12
*
20
*
10
*
Estradiol [ng/ml]
*
Estradiol [ng/ml]
15
8
*
6
10
*
4
5
*
2
0
0
ctr-
MOI15
MOI50
MOI100 MOI500
6h
24h
induced by E1A adenoviral early protein to facilitate viral replication. Adrenal tropism and lytic activity of wild-type adenovirus represent beneficial characteristics that could be exploited to engineer oncolytic viruses against adrenocortical carcinoma.
At variance with the wild-type adenovirus, transduction with replication-deficient AdVs at low titers had no significant cytopathic effects on adrenocortical cells. But, at high titers,
simulating the viral dose which might be received by cells after in vivo systemic AdV administration for gene therapy, AdV infection led to cytopathic changes, decreased cell proliferation, and provoked G2/M-arrest. These results in human adrenocortical cell lines are at variance with observations performed on primary cell cultures obtained from bovine adrenal cortex, which increased their proliferation following AdV infection (Alesci et al., 2002), but are in agreement with the
a
25
ctr-
b
120
Adnull
StAR mRNA [x10 copies]
*
*
AdEGFP
20
100
*
Ad5
LPS
DAX-1 mRNA
80
15
CpG
*
60
10
*
*
*
40
*
*
5
20
*
*
0
6h
24h
0
6h
24h
C
30
d 300
*
*
*
SF-1 mRNA [x102 copies]
25
250
*
20
*
*
*
*
CYP21 mRNA
200
*
15
*
150
*
*
*
*
*
10
100
5
50
I
0
6h
24h
0
6h
24h
e
30
f
*
80
CYP19 mRNA [x103 copies]
25
70
*
20
*
*
CYP11B1 mRNA
60
50
15
40
*
*
10
30
*
*
*
*
20
*
5
*
*
10
I
I
0
0
6h
24h
6h
24h
g
40
h
35
*
Cholesterol
StAR
CYP17
SULT2A1
CYP11B2 mRNA
30
CYP11A
17 OH Preg
DHEA
DHEA
*
Pregnenolone
sulfate
25
HSD382
20
Progesterone
17 OH Prog
Androstenedione
CYP21
HSD 1783
15
*
*
Deoxycorticosterone
Deoxycortisol
Testosterone
10
*
CYP11B1
CYP19
Corticosterone
5
CYP112
CORTISOL
ESTRADIOL
0
ALDOSTERONE
6h
24h
28
60
ctr-
Adnull
24
50
*
AdEGFP
Ad5
20
TLR9 mRNA
IRF7 mRNA
40
CpG
16
LPS
30
*
12
20
8
*
4
10
*
*
*
0
0
6h
24h
6h
24h
3000
60000
INF-a mRNA
2500
50000
2000
INF-B mRNA
40000
1500
30000
1000
20000
500
10000
0
0
6h
24h
6h
24h
30000
8000
25000
INFAR-1 mRNA
7000
20000
INFAR-2 mRNA
6000
5000
15000
4000
10000
3000
5000
2000
1000
0
0
6h
24h
6h
24h
30
8
*
25
*
7
*
*
TNF-a mRNA
6
IL-6 mRNA
20
5
15
*
4
*
*
10
3
*
*
*
*
2
*
5
1
0
0
6h
24h
6h
24h
cytopathic effects induced by E1/E3-deleted AdVs in other cell types, for example, hepatocytes (Brand et al., 1999) and airway epithelial cells (Hartmann et al., 2007). Induction of phenotypic changes was not due to transgene expression, but probably to adenoviral structural components or residual gene expression (Walters et al., 2002), since they were observed also with the AdV carrying no transgene. The NCI-H295R cells showed higher susceptibility to cytopathic effects than SW-13 cells. This could be related to different features of the two tumor cell lines: the SW-13 cell line was established from a more aggressive form of adrenocortical carcinoma mutated in p53 gene and has lost the ability to produce hormones, whereas NCI-H295R cells contain wild-type p53, produce steroid hormones and retain slow growth rate.
With particular attention, we studied the effect of AdVs on steroidogenesis. Steroid hormone production was induced in MOI-dependent manner by both AdVs and Ad5WT and the effect was not related to the transgene. Moreover, treatment with LPS and CpG-ODN had no marked effect on hormone production, indicating the effect was adenovirus-specific. Since increased steroid hormone production was already observed at 6 h pi, this effect might be related to adenovirus entry and/or its early gene expression. Indeed, recent studies using adenovirus empty capsids strongly suggest that cellular innate immune response to AdV infection is most likely provoked by early steps of virion infection, regardless of modifications to vector genome (Stilwell and Samulski, 2004). As observed in cells of the immune system (Benihoud et al., 2007), adrenocortical cells responded to adenoviral infection with increased expression of the inflammatory cytokines TNF-& and IL-6, which represent important players in the initial nonspecific immune response to viral infection, and with increased expression of IRF7 mRNA, an important component of cellular IFN response. Besides with cytokine response, human adrenocortical cells participated in the host response against adenoviral infection with an acute stress response characterized by markedly increased secretion of cortisol and other steroid hormones. Steroid hormone hypersecretion was related to a rapid upregulation of key steroidogenic enzymes and regulators of steroidogenesis. In particular, expression of StAR was significantly induced by both AdVs and Ad5WT. StAR is a global activator of steroidogenesis that facilitates uptake and intramitochondrial transfer of cholesterol to be further metabolized into steroid hormone precursors. Expression of the global negative regulator DAX-1 and its antagonist SF-I was also affected accordingly, with a marked downregulation of DAX-I and upregulation of SF-I at the early times pi. The ratio between DAX-I and SF-I influences the expression of StAR and steroidogenic enzymes. Consistent with SF-I and StAR activation, expression of CYP21 and CYP19 enzymes was significantly upregulated early pi, whereas mRNAs of CYPI IBI and CYPI IB2, which catalyze the last steps of aldosterone and cortisol production, respectively, displayed a marked upregulation at 24h pi.
We think induction of steroidogenesis, and in particular of cortisol production, could represent a general mechanism of response of the adrenocortical cell to viral infection. This is suggested by our analysis of human adrenocortical cell response to other viruses, such as human cytomegalovirus (HCMV). In this regard, we recently demonstrated HCMV infects and replicates in human adenocortical cells and induces steroid hormone production within the first hour pi. Moreover, StAR and steroidogenic enzyme expression, as well as cytokine expression, was induced in HCMV infected cells with a pattern very similar to that observed in adenovirus infected cells (Trevisan M. and Barzon L., unpublished observations). Thus, increased cortisol production and inflammatory cytokine expression could represent the innate immune-endocrine response to viral infection of adrenocortical cells. Although TLR-9 expression was low in our model, it cannot be excluded
that this response is mediated by TLR-dependent pathways, since IRF7 expression was induced after infection (Zhu et al., 2007).
In conclusion, this study demonstrates AdVs efficiently transduce adrenocortical cells in vitro. At low-doses, AdVs did not show any cytopathic change and had mild effects on steroidogenesis. Low vector doses could reach the adrenal gland when AdVs are directly injected into the target tissue (e.g., intratumor injection). The problem might arise in the case of systemic vector administration, when the injected viral dose must be two- to fivefold higher to reach the target tissue at sufficient therapeutic concentration, due to dissemination of particles in different organs. At higher doses, AdVs and wild type adenovirus infection provoked a marked increase of steroid hormone production, especially of cortisol. This effect was related to induction of adrenal StAR and steroidogenic enzyme expression, probably as a response to adenoviral entry and/or early phases of infection. In the context of adenoviral vector use for gene therapy, this effect could be beneficial because of the anti-inflammatory and immunosuppressive effects of glucocorticoids at high concentrations, which could counteract host immune-inflammatory response to AdV or oncolytic adenovirus administration, even though pro- inflammatory effects have also been suggested (Arafah, 2006). The results of this study contribute to our understanding of virus-host interaction during natural adenovirus infection and in clinical trials of gene therapy using AdVs. Moreover, our findings further support the role of the adrenal cortex in host innate immune-endocrine response to infectious agents.
Acknowledgments
This work was supported by grant no. 2004063894_003 from MIUR (Ministero dell’istruzione dell’Università e della Ricerca) to Luisa Barzon and by grant no. RSF 271/07 from Veneto Region to Giorgio Palù.
Literature Cited
Alesci S, Ramsey WJ, Bornstein SR, Chrousos GP, Hornsby PJ, Benvenga S, Trimarchi F, Ehrhart-Bornstein M. 2002. Adenoviral vectors can impair adrenocortical steroidogenesis: Clinical implications for natural infections and gene therapy. Proc Natl Acad Sci USA 99:7484-7489.
Arafah BM. 2006. Hypothalamic pituitary adrenal function during critical illness: Limitations of current assessment methods. J Clin Endocrinol Metab 91:3725-3745.
Barzon L, Boscaro M, Palù G. 2004. Endocrine aspects of cancer gene therapy. Endocr Rev 25:1-44.
Barzon L, Masi G, Pacenti M, Trevisan M, Fallo F, Remo A, Martignoni G, Montanaro D, Pezzi V, Palù G. 2008. Expression of aromatase and estrogen receptors in human adrenocortical tumors. Virchows Arch 452:181-191.
Benihoud K, Esselin S, Descamps D, Julienne B, Salone B, Bobé P, Bonardelle D, Connault E, Opolon P, Saggio I, Perricaudet M. 2007. Respective roles of TNF& and IL-6 in the immune response-elicited by adenovirus-mediated gene transfer in mice. Gene Ther 14:533-544. Bornstein SR, Rutkowski H, Vrezas I. 2004a. Cytokines and steroidogenesis. Mol Cell Endocrinol 215:135-141.
Bornstein SR, Zacharowski P, Schumann RR, Barthel A, Tran N, Papewalis C, Rettori V, McCann SM, Schulze-Osthoff K, Scherbaum WA, Tarnow J, Zacharowski K. 2004b. Impaired adrenal stress response in Toll-like receptor 2-deficient mice. Proc Natl Acad Sci USA 101:16695-16700.
Brand K, Klocke R, Possling A, Paul D, Strauss M. 1999. Induction of apoptosis and G2/M arrest by infection with replication-deficient adenovirus at high multiplicity of infection. Gene Ther 6:1054-1063.
Cerullo V, Seiler MP, Mane V, Brunetti-Pierri N, Clarke C, Bertin TK, Rodgers JR, Lee B. 2007. Toll-like receptor 9 triggers an innate immune response to helper-dependent adenoviral vectors. Mol Ther 15:378-385.
Cutlip RC, Mcclurkin AW. 1975. Lesions and pathogenesis of disease in young calves experimentally induced by a bovine adenovirus type 5 isolated from a calf with weak calf syndrome. Am J Vet Res 36:1095-1098.
Gazdar AF, Oie HK, Shackleton CH, Chen TR, Triche TJ, Myers CE, Chrousos GP, Brennan MF, Stein CA, La Rocca RV. 1990. Establishment and characterization of a human adrenocrtical carcinoma cell line that expresses multiple pathways of steroid biosynthesis. Cancer Res 50:5488-5496.
Groot-Wassink T, Aboagye EO, Glaser M, Lemoine NR, Vassaux G. 2002. Adenovirus biodistribution and non-invasive imagin of gene expression in vivo by positron emission tomography using the human sodium/iodide symporter as a reporter gene. Human Gene Ther 13:1723-1735.
Hartmann ZC, Black EP, Amalfitano A. 2007. Adenoviral infection induces a multi-faced innate cellular immune response that is mediated by the toll-like receptor pathway in A549 cells. Virology 358:357-372.
ADRENOCORTICAL RESPONSE TO ADENOVIRAL INFECTION
Hoenig EM, Margolis G, Kilham L. 1974. Experimental adenovirus infection of the mouse adrenal gland. Il. Electron microscopic observations. Am J Pathol 75:375-394.
Leibovitz A, McCombs WM III, Johnston D, McCoy CE, Stinson JC. 1973. New human cancer cell culture lines. I. SW-13, small-cell carcinoma of the adrenal cortex. J Natl Cancer Inst 51:691-697.
Liu Q, Muruve DA. 2003. Molecular basis of the inflammatory response to adenovirus vectors. Gene Ther 10:935-940.
Lowe SW, Ruley HE. 1993. Stabilization of the p53 tumor suppressor is induced by adenovirus 5 EIA and accompanies apoptosis. Genes Dev 7:535-545.
Margolis G, Kilham L, Hoenig EM. 1974. Experimental adenovirus infection of the mouse adrenal gland. I. Light microscopic observations. Am J Pathol 75:363-374.
Medvedev N, Shastina GV. 1978. Morphology of the adrenal cortex in infants with generalized adenovirus infections. Arkh Patol 40:32-36.
Mickelson CA. 2000. Department of Health and Human Services National Institutes of Health Recombinant DNA Advisory Committee. Minutes of meeting March 8-10,2000. Hum Gene Ther 11:2156-2192.
Rainey WE, Bird IM, Mason JI. 1994. The NCI-H295 cell line: A pluripotent model for human adrenocrtical studies. Mol Cell Endocrinol 100:45-50.
Rainey WE, Saner K, Schimmer BP. 2004. Adrenocortical cell lines. Mol Cell Endocrinol 228:23-38.
Shayakhmetov DM, Li ZY, Ni S, Lieber A. 2004. Analysis of adenovirus sequestration in the liver, transduction of hepatic cells, and innate toxicity after injection of fiber-modified vectors. J Virol 78:5368-5381.
Shayakhmetov DM, Gaggar A, Ni S, Li ZY, Lieber A. 2005. Adenovirus binding to blood factors results in liver cell infection and hepatotoxicity. J Virol 79:7478-7491.
Stilwell JL, Samulski RJ. 2004. Role of viral vectors and virion shells in cellular gene expression. Mol Ther 9:337-346.
Strauber E, Card C. 1978. Experimental intraamniotic exposure of bovine fetuses with subgroup 2, type 7 adenovirus. Can J Compar Med 42:466-472.
Thomas CE, Ehrhardt A, Kay MA. 2003. Progress and problems with the use of viral vectors for gene therapy. Nat Rev Genet 4:346-358.
Tran N, Koch A, Berkels R, Boehm O, Zacharowski PA, Baumgarten G, Knuefermann P, Schott M, Kanczkowski W, Bornstein SR, Lightman SL, Zacharowski K. 2007. Toll-like receptor 9 expression in murine and human adrenal glands and possible implications during inflammation. J Clin Endocrinol Metab 92:2773-2783.
Walters RW, Freimuth P, Moninger TO, Ganske I, Zabner J, Welsh MJ. 2002. Adenovirus fiber disrupts CAR-mediated intercellular adhesion allowing virus escape. Cell 110:789-799. Wang Y, Groot-Wassink T, Lemoine NR, Vassaux G. 2003. Cellular characterization of the tropism of recombinant adenovirus for the adrenal glands. Eur J Clin Invest 33:794-798. Zacharowski K, Zacharowski PA, Koch A, Baban A, Tran N, Berkels R, Papewalis C, Schulze- Osthoff K, Knuefermann P, Zähringer U, Schumann RR, Rettori V, McCann SM, Bornstein SR. 2006. Toll-like receptor 4 pays a crucial role in the immune-adrenal response to systemic inflammatory response syndrome. Proc Natl Acad Sci USA 103:6392-6397.
Zhu J, Huang X, Yang Y. 2007. Innate immune response to adenoviral vectors is mediated by both toll-like receptor-dependent and-independent pathways. J Virol 81:3170-3180.