Long-term infection passaging of Human Adenovirus 36 in monkey kidney cells

ABSTRACT

Human Adenovirus 36 (HAdV-36) has been related to diverse effects on metabolism and may attenuate the lipid accumulation in kidneys with increased adiposity. Some of these effects would be related to viral persistence. However, until now, a model of persistent in vitro infection by HAdV-36 is unknown. In this study, we examined the cells of the Vero lineage to explore their permissiveness to long-term HAdV-36 infection. HAdV-36 was productively replicated in Vero cells and maintained long-term infection for up to 35 cell passages. A subculture was obtained from the cells that survived the primary infection at a low MOI (0.5). The production of the extracellular infectious virus with titers ranging from 10⁴ to 10⁶ TCID₅₀/mL and DNA-bearing cells was detected. In long-term infected cells, the intracellular distribution of viral antigen was demonstrated by performing immunolocalization (IFI) and expression of cell-viral antigen in 50% of cells by flow cytometry, using anti-HAdV-36 hyperimmune rabbit serum. Furthermore, E1a and E4orf1 genes in long-term infected passages showed a decreasing trend. Our preliminary results reveal that renal epithelial monkey cells are permissive for the productive infection of HAdV-36. Vero cell culture long-term infection might be a promising model for addressing the fundamental aspects of the HAdV-36 biology that cannot reveal broadly-used cultures, which do not maintain long-term infection in primary or transformed cells.

HAdV-36; Ad36; Vero cells; HAdV-36 long-term in vitro infection

INTRODUCTION

HAdVs were first recognized and isolated as pathogens approximately 70 years ago. HAdVs are non-enveloped dsDNA viruses that belong to the Adenoviridae family and the Mastadenovirus genus. Over 80 types have been described and grouped into seven species (A-G). Species D contains the most types, including HAdV-36, first described in 1980, with genomic characterization in 2010¹ . The HAdV-36 E4orf1 protein has been proposed as a biological factor that can improve glucose and lipid metabolism, even with weight gain and adiposity² . Furthermore, E4orf1 might attenuate the lipid accumulation in kidneys³ .

Additionally, HAdV-36 can spread to most organs (liver, kidneys, lungs, brain, and adipose tissue) of naturally or experimentally infected animals⁴ , ⁵ , which suggests a link to HAdV-36 persistence⁶ . However, the molecular mechanism through which HAdV-36 can establish a persistent infection is unknown. The viral replication and capacity for adenovirus production have been examined using renal epithelial monkey cell culture (Vero cell lineage)⁷ .

MATERIALS AND METHODS

HAdV-36 (VR-1610) was obtained from ATCC (Manassas, VA, USA). The stock of HAdV-36 was purified, propagated and titrated as described by Dhurandhar et al. ⁸ A viral stock at a 10¹² TCID₅₀/mL concentration was used in the experiments. Vero cells (ATCC CCL-81) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 6% of fetal bovine serum (FBS) under standard conditions.

Cell culture infection: acute and long-term infected cultures

Confluent monolayers of Vero cells were inoculated with HAdV-36 stock at a multiplicity of infection (MOI) of 1.0 for acute infection and 0.5 for long-term infection. Both were grown in DMEM with 2% FBS in T25 flasks in triplicate. The acute infection was considered until 14 days post-inoculation (dpi). On the other hand, for long-term infected cultures, incubation was maintained for eight days, and then the supernatant (SN) was removed and replaced with a fresh culture medium. The incubation continued for another seven days. After this incubation, the surviving infected cells formed confluent monolayers (at 20 dpi) in order to obtain long-term infected cultures that were subcultured (henceforth, passaged) usually every week for 35 passages, which is equivalent to eight months. In a similar way, non-infected Vero cells, used as control, were incubated. Two-thirds of these were used to determine the cells bearing viral DNA, cells expressing viral antigen and the mRNA expressing E1a and E4orf1 genes. Also, the final third was subcultured.

TCID50 assay

We analyzed HAdV-36 infectivity or productive infection in the passages based on the production of extracellular virions released in the SNs using a TCID₅₀ assay. As described briefly by Dhurandhar et al. ⁸ , four wells of 96 well-plate were inoculated with a ten-fold dilution series, and the Spearman–Karber’s method⁹ was used to calculate the viral titer.

HAdV-36 DNA detection

In the long-term infected passages cells, the infection by HAdV-36 was proven using conventional PCR. The primers used were F (5´-TGAGCAGCAGATGGCTCTAATCTC-3´) and R (5´-GGTCTTCTTCTGAGGGTGATGACTC-3´), as previously described¹⁰ , to amplify a fragment of the E1a gene yielding an amplicon size of approximately 320 bp. Non-infected and acutely infected cells were included as controls. An Applied Biosystems Veriti thermocycler (CA, USA) was used. The PCR products were separated by agarose gel electrophoresis (1.5%), visualized using ethidium bromide staining (0.5 µg/mL) and analyzed using a 212 PRO Carestream photo documentation system (Carestream 212 PRO, Gel Logic, Rochester, USA). The amplification products of PCR were purified with a QIAquick PCR Purification Kit (Qiagen, Germany) and sequenced using a commercial supplier to confirm the genetic identity of the virus.

For antigen detection by immunostaining assays, a hyperimmune rabbit serum against HAdV-36 (from now on referred to as pAb) was produced in two New Zealand white rabbits by immunization with concentrated and purified HAdV-36, following the protocols described by Sarmiento et al. ¹¹ .

Viral antigen detection by indirect immunofluorescence (IFI)

Cells from long-term infected passages were trypsinized, seeded and grown until they reached the confluence on the wells chamber slides (Thermo Fisher Scientific, Waltham, MA, USA). Similarly, non-infected and acutely infected cells were seeded as controls. Fluorescence staining was performed as described by Sarmiento et al. ¹¹ with slight adaptations. We used pAb (1:300 in blocking solution) as the primary antibody and fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit IgG (H+L) (Jackson ImmunoResearch, W. Baltimore Pike, USA) (1:500 in PBST 0.1%) as the secondary antibody. Nuclei were counterstained with DAPI (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA). Moreover, the cells were viewed using a fluorescence microscope (Nikon, Eclipse E600, Japan).

Viral antigen detection by flow cytometry

We used 5 × 10⁵ cells of cultures of the long-term infected, acutely infected and non-infected to detect the viral antigen. We used pAb diluted 1:300 in PBST 0.1% as the primary antibody and fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit IgG (H+L) (Jackson ImmunoResearch, W. Baltimore Pike, USA) (1:500 in PBST 0.1%) as the secondary antibody. Subsequently, the cells were resuspended in one mL of PBS and 1 × 10⁴ cells were analyzed using flow cytometry. Data were then collected using a FACSCanto II cytometer (Becton Dickinson, San Jose, CA, USA) and analyzed using the FACSDIVA software (version 6.1.3, Becton Dickinson). For each sample, at least 10,000 events were acquired. Fluorochrome‐labeled isotype‐matched control pAb was used to determine background staining.

Expression of E1A and E4orf1 mRNA in long-term cultures

Cells were washed three times with PBS and lysed in Trizol reagent (Thermo Fisher Scientific, Waltham, MA, USA). The total RNA extracted after treatment with DNase was used to synthesize cDNA using Oligo dT and MMLV RT (Promega, Madison, WI, USA). The amplification was done using the SYBR Green (Thermo Fisher Scientific, Waltham, MA, USA) and QIAquant 96 systems (Qiagen Ca, USA); the primers used were: α-tubulin (Fw 5ꞌ-CAGATGCCAAGTGACAAGAC-3ꞌ Rv 5ꞌ-ACTCCAGCTTGGACTTCTTG-3ꞌ), E1A (Fw 5ꞌ AGATGCCCATGATGACGACC-3ꞌ Rv 5ꞌ GCCTTGTACTGGCTCGGTAG3ꞌ) and E4ORF1 (Fw 5ꞌ CCAGGAAGGGGCTAGCAATA3ꞌ Rv 5ꞌ AATCACTCTCTCCAGCAGCAGG3ꞌ). Relative gene expression was calculated using the comparative ∆∆Ct method (where C_T is the threshold cycle) and was used to determine the fold increase in gene expression of long-term infected cells and acutely infected cells normalized to α-tubulin expression.

RESULTS AND DISCUSSION

In this study, the infection of Vero cells by HAdV-36 resulted in productive infection, with viral DNA-bearing cells and expression of the HAdV-36 antigen in almost 50% of infected cells in the long term, from 20 dpi to 200 dpi or higher, which is consistent with a persistent carrier-state infection.

The persistence of HAdV-36 is plausible based on Dhurandhar et al. ⁵ and Krishnampuram et al .⁴ , whose independent studies showed that experimentally infected monkeys and mice harbored the HAdV-36 in the liver, lungs, and kidneys for at least four months after inoculation.

Using Vero cells that mimic the entry cells of adenoviruses, we managed to replicate HAdV-36 in them, as was done for HAdV-5⁷. Furthermore, we obtained long-term infected cultures by infecting Vero cells with a low MOI (0.5). Similarly, Tirado et al. ¹² used this to establish a persistently infected culture. Our control of the infection acute phase at eight dpi allowed the surviving cells to multiply until the confluence at 14 dpi. For instance, the study by Alidjinou et al .¹³ used this strategy to establish persistent infection by Coxsackievirus in vitro .

Afterwards, we performed the first passage and obtained a 10¹⁰ TCID₅₀/mL viral titer. Nevertheless, we found that some cells grew with this level of the infectious virus and multiplied to the confluence at 20 dpi. Similarly, in research by Kopliku et al. ¹⁴ , this strategy allowed the establishment of a model of persistent infection in bovine kidney epithelial cells with FMDV.

In other words, the strategies mentioned allowed for a balance between the presence of the virus and cell multiplication, which are characteristics of viral persistence in culture¹⁵ .

Additionally, in the subsequent passage at 20 dpi, we obtained confluent monolayers 72h after the passage; this phenomenon occurred in the following passages, unlike non-infected cells that reached confluence between 24 to 48 h. In this sense, it has been shown that surviving cells that are regularly passaged for a long time become steady-state persistently infected cells¹⁶ .

Notably, we assessed the infectivity of HAdV-36 in long-term infected passages quantifying an extracellular infectious virus. From 20 dpi until after 200 dpi, cultures continuously released infectious viruses with titers ranging from 10⁴ to 10⁶ TCID50/mL, almost half of the titer obtained in the infection at 14 dpi (10¹⁰ TCID₅₀/mL) ( Figure 1A ). These results are interpreted as active viral replication, as revealed by the infectivity that has been evaluated in other in vitro persistence models¹³ . Indeed, we chose it here because other methods, such as PCR, have low specificity for detecting infectious viruses¹⁷ . However, we used PCR to obtain a 320 bp amplicon, corresponding to a fragment of the E1a gene, as an indicator of HAdV-36 entry into Vero cells ( Figure 1B ). E1A was tested because these immediate-early proteins drive cells into the S phase and induce the expression of all other early viral genes. Thus, we viewed positive signals for the HAdV-36 antigen using immunostaining-IFI ( Figure 2, A to I ). However, the cells did not show a homogeneous pattern in viral antigen expression (cells with different levels of green fluorescence brightness) ( Figure 2 , F and I ). Therefore, the quantitative expression of the viral antigen was determined using flow cytometry assays and the expression antigen was found in ~50% of the long-term infected cells ( Figure 2L ), which was lower than the expression in the acutely infected cultures (~80%) ( Figure 2K ).

Figure 1
The persistent presence of HAdV-36 in long-term infected cultures: A) In long-term infected passages (more than 20 dpi), the production of extracellular virions was found with titers ranging between 104 to 106 TCID50/mL, which were lower than the titer found in acutely infected cultures (1010.5 TCID50/mL at 14 dpi or less). Representative experiments throughout the passages; B) Presence of HAdV-36 DNA-bearing cells in long-term infected cultures were revealed by obtaining the 320 bp length amplicons corresponding to a fragment of the E1a gene using conventional PCR. Representative experiments of acutely infected cells at 14 dpi, in lane 6, and long-term infected passages at 53-, 110-, 184-, 216- and 244-days post-inoculation, respectively, in lanes 7-11.

Figure 2
The HAdV-36 antigen was expressed in long-term infected Vero cells. Non-infected cells as control (A, B, C). Immunostaining of antigen in acutely infected cells at 14 days post-inoculation (D, E, F). Representative images of immunostaining for HAdV-36 antigen in long-term infected cells (216 days post-inoculation) (G, H, I) show the pattern of positive immunolocalization of the viral antigen (green fluorescence). Nuclei were counterstained with 4´, 6-diamidino-2-phenylindole (blue fluorescence). Microphotographs at 400X magnification. Images were captured using fluorescence microscopy. Scale bar 50 µm; Quantitative expression of HAdV-36 antigen in Vero cell cultures. Long-term infected Vero cell cultures expressed the viral antigen in 50% of the cells (L) compared to acutely infected cells in which 80% expressed the viral antigen (K), analyzed through flow cytometry. Non-infected cells were included as control cells (J). Representative experiments are shown at 14- and 216-days post-inoculation.

As mentioned earlier, the carrier-state persistence feature shows a heterogeneous pattern of virus-infected cells¹⁵ . Alternatively, it has been shown that there is an unstable dynamic phase at an early stage of the process to establish a persistent infection in a steady-state in cultured cells, for which the viruses capable of establishing a persistent infection would be selected¹⁶ , especially to control Vero cell infection by HAdV-36 for a longer time.

Long-term infections have been documented in other adenovirus species and other cell lineages¹⁸ . However, none of these infections have been with HAdV-36. In addition, we have achieved long-term infections in murine macrophages in our lab, but they have not been sustained as much as in Vero cells.

In another sense, several studies illustrate that E4orf1 and E1A are essential for viral replication and interaction with each other. In addition, these proteins have been related to the reprogramming effect of cellular metabolism caused by adenovirus infection² . Consequently, we analyze the relative expression of E1a and E4orf1 in HAdV-36 long-term infection versus acute infection using α-tubulin as a housekeeping gene ( Figure 3 ). In long-term infected cells, we found low relative expression of E1a ( Figure 3A ) and a decrease in the relative expression of E4orf1 after 53 dpi ( Figure 3B ). However, we did not find a significant difference in gene mRNA ratios in acutely and long-term infected cell passages beyond 53 dpi. In contrast, this was observed on an in vitro model of persistent HAdV-C5 infection (in interferon-expressing cells) that resulted in the repression of E1a expression¹⁹ .

Figure 3
Expression of E1a and E4orf1 were induced at detectable levels since the first long-term infected passages, in decreasing tendency. Relative expression of E1a (A) and E4orf1 (B) mRNA was determined in passages of long-term and acutely infected Vero cells through normalization of the amount of mRNA of α-tubulin in the housekeeping gene qRT-PCR using the ΔΔ Ct method. Representative experiments are shown at 14-, 53-, 110-, and 244-days post-inoculation. The data represent the average of three independent experiments. Significant differences for α-tubulin are indicated by (*) p < 0.05.

Interestingly, the expression of E4orf1 mRNA in the long-term infected passage at 53 dpi showed a significant difference, although, after this time, the expression was not significant with respect to acutely infected cells ( Figure 3B ). This finding leads to further investigation of whether E4orf1 can stimulate the secretion of growth factors that subvert apoptosis, as found by Jerebtsova et al .²⁰ in renal glomerular endothelial cells, knowing that the mechanisms that favor cell survival play an essential role in establishing viral persistence. The relative expression level of both viral genes was lower than the expression level of the housekeeping gene. Equally, Dickherber and Garnett-Benson²¹ studied this in their model of persistent infection by Ad5dl309 in lymphocytes. Moreover, in persistently infected cells, it has been found that viral gene transcription can be repressed. On the other hand, the viral gene expression pattern has been studied in murine fibroblasts through an in vitro HAdV-36 infection in the short term (1–2 weeks post-infection)¹⁰ , ²² . However, nothing is known about expression during long-term infection in renal epithelial cells. Indeed, our findings suggest that HAdV-36 was established as a persistent carrier-state infection in Vero cells.

CONCLUSION

Finally, our results prelude investigating genes involved in HAdV-36 persistence in epithelial cells. In addition, we could have a model to study the effect of HAdV-36 and its proteins on the expression of molecules related to the entry of LCFA, such as CD36 and FATP4, potential targets to attenuate lipid accumulation in the kidneys.

ACKNOWLEDGMENTS

The authors are grateful to Dr. Alejandro L Saavedra, Unidad de Aplicaciones Avanzadas en Microscopia del Instituto Nacional de Cancerologia (INCan), la Red de Apoyo a la Investigacion (RAI-UNAM), and Dr. Marco Alonso Andonegui Elguera, Subdireccion de Investigacion Basica, INCan. CDMX, Mexico. We thank HT Hector Trinidad Bibiano, Division de Anatomia Patologica del Hospital General, Dr. Manuel Gea Gonzalez, CDMX, and MsC Enrique Sanchez Flores for helpful technical advice. We also thank Dr. Rocio Tirado Mendoza, Facultad de Medicina, UNAM, for providing a cell Vero lineage and valuable recommendations for managing cell cultures. PAV thanks CONACYT-Mexico for awarding the doctoral fellowship (Nº 792588). This work is part of the PhD dissertation submitted by Patricia Alarcón-Valdes in partial fulfillment of the degree requirements.

REFERENCES

• ¹
  Arnold J, Jánoska M, Kajon AE, Metzgar D, Hudson NR, Torres S, et al. Genomic characterization of human adenovirus 36, a putative obesity agent. Virus Res. 2010;149:152-61.
• ²
  Hegde V, Na HN, Dubuisson O, Burke SJ, Collier JJ, Burk D, et al. An adenovirus-derived protein: a novel candidate for anti-diabetic drug development. Biochimie. 2016;121:140-50.
• ³
  Afruza R, Akheruzzaman M, Dhurandhar NV, Hegde V. E4orf1, an Adeno-viral protein, attenuates renal lipid accumulation in high fat fed mice: a novel approach to reduce a key risk factor for chronic kidney disease. Heliyon. 2020;6:e05261.
• ⁴
  Krishnapuram R, Kirk-Ballard H, Zuberi A, Dhurandhar NV. Infectivity period of mice inoculated with human adenoviruses. Lab Anim. 2011;45:103-8.
• ⁵
  Dhurandhar NV, Whigham LD, Abbott DH, Schultz-Darken NJ, Israel BA, Bradley SM, et al. Human adenovirus Ad-36 promotes weight gain in male rhesus and marmoset monkeys. J Nutr. 2002;132:3155-60.
• ⁶
  Hwang KA, Park S, Ahn JH, Nam JH. Development of a standard protocol for quantitative polymerase chain reaction to detect adenovirus 36, which is associated with obesity. Acta Virol. 2018;62:350-9.
• ⁷
  Lee DK, Park J, Seo DW. Suspension culture of Vero cells for the production of adenovirus type 5. Clin Exp Vaccine Res. 2020;9:48-55.
• ⁸
  Dhurandhar NV, Israel BA, Kolesar JM, Mayhew G, Cook ME, Atkinson RL. Transmissibility of adenovirus-induced adiposity in a chicken model. Int J Obes Relat Metab Disord. 2001;25:990-6.
• ⁹
  Ramakrishnan MA. Determination of 50% endpoint titer using a simple formula. World J Virol. 2016;5:85-6.
• ¹⁰
  Rathod M, Vangipuram SD, Krishnan B, Heydari AR, Holland TC, Dhurandhar NV. Viral mRNA expression but not DNA replication is required for lipogenic effect of human adenovirus Ad-36 in preadipocytes. Int J Obes (Lond.). 2007;31:78-86.
• ¹¹
  Sarmiento RE, Tirado RG, Valverde LE, Gómez-Garcia B. Kinetics of antibody-induced modulation of respiratory syncytial virus antigens in a human epithelial cell line. Virol J. 2007;4:68.
• ¹²
  Tirado R, Sarmiento RE, Gómez B. Spontaneous shift of an HSV type 1 productive, persistently infected macrophage-like culture to a non-productive culture. Res Virol. 1998;149:145-51.
• ¹³
  Alidjinou EK, Engelmann I, Bossu J, Villenet C, Figeac M, Romond MB, et al. Persistence of Coxsackievirus B4 in pancreatic ductal-like cells results in cellular and viral changes. Virulence. 2017;8:1229-44.
• ¹⁴
  Kopliku L, Relmy A, Romey A, Gorna K, Zientara S, Bakkali-Kassimi L, et al. Establishment of persistent foot-and-mouth disease virus (FMDV) infection in MDBK cells. Arch Virol. 2015;160:2503-16.
• ¹⁵
  Nathanson N. Viral pathogenesis and immunity. 2^nded. Amsterdam: Academic Press; 2007.
• ¹⁶
  Ito M, Takeuchi T, Nishio M, Kawano M, Komada H, Tsurudome M, et al. Early stage of establishment of persistent Sendai virus infection: unstable dynamic phase and then selection of viruses which are tightly cell associated, temperature sensitive, and capable of establishing persistent infection. J Virol. 2004;78:11939-51.
• ¹⁷
  Funk DJ, Bullard J, Lother S, Grande GV, Garnett L, Doan K, et al. Persistence of live virus in critically ill patients infected with SARS-COV-2: a prospective observational study. Crit Care. 2022;26:10.
• ¹⁸
  Chu Y, Sperber K, Mayer L, Hsu MT. Persistent infection of human adenovirus type 5 in human monocyte cell lines. Virology. 1992;188:793-800.
• ¹⁹
  Zheng Y, Stamminger T, Hearing P. E2F/Rb family proteins mediate interferon induced repression of Adenovirus immediate early transcription to promote persistent viral infection. PLoS Pathog. 2016;12:e1005415.
• ²⁰
  Jerebtsova M, Kumari N, Obuhkov Y, Nekhai S. Adenoviral E4 gene stimulates secretion of pigmental epithelium derived factor (PEDF) that maintains long-term survival of human Glomerulus-derived endothelial cells. Mol Cell Proteomics. 2012;11:1378-88.
• ²¹
  Dickherber ML, Garnett-Benson C. NAD-linked mechanisms of gene de-repression and a novel role for CtBP in persistent adenovirus infection of lymphocytes. Virol J. 2019;16:161.
• ²²
  Krishnapuram R, Dhurandhar EJ, Dubuisson O, Hegde V, Dhurandhar NV. Doxycycline-regulated 3T3-L1 preadipocyte cell line with inducible, stable expression of adenoviral E4orf1 gene: a cell model to study insulin-independent glucose disposal. PLoS One. 2013;8:e60651.

Publication Dates

• Publication in this collection
  14 Nov 2022
• Date of issue
  2022

History

• Received
  24 May 2022
• Accepted
  19 Sept 2022