An Ex Vivo Chicken Primary Bursal-cell Culture Model to Study Infectious Bursal Disease Virus Pathogenesis

Infectious bursal disease virus (IBDV) is a birnavirus of economic importance to the poultry industry. The virus infects B cells, causing morbidity, mortality, and immunosuppression in infected birds. In this study, we describe the isolation of chicken primary bursal cells from the bursa of Fabricius, the culture and infection of the cells with IBDV, and the quantification of viral replication. The addition of chicken CD40 ligand significantly increased cell proliferation fourfold over six days of culture and significantly enhanced cell viability. Two strains of IBDV, a cell-culture adapted strain, D78, and a very virulent strain, UK661, replicated well in the ex vivo cell cultures. This model will be of use in determining how cells respond to IBDV infection and will permit a reduction in the number of infected birds used in IBDV pathogenesis studies. The model can also be expanded to include other viruses and could be applied to different species of birds.


Introduction
The global poultry industry is essential to secure enough food for an expanding human population. However, immunosuppression is the threat to food security and the welfare of affected birds and represents a key economic challenge to the industry. The majority of cases of immunosuppression in chickens are caused by the infection with immunosuppressive viruses, with those most responsible for impairing acquired immunity having a tropism for T and B lymphocytes 1 . In birds, the majority of B cells are located within an organ known as the bursa of Fabricius (BF). B cells are susceptible to infection with several immunosuppressive viruses, including those that cause lysis, such as IBDV and Marek's disease virus (MDV), and those that cause transformation, such as avian leukosis virus (ALV) and reticuloendotheliosis virus (REV).
In order to develop better strategies for controlling these infections, it is essential to characterize the interaction of the viruses with chicken B cells. However, when B cells are removed from a bird, they do not survive well in ex vivo culture 2 , making it difficult to perform a thorough analysis of the interactions of IBDV with chicken B cells, or the early events following ALV or REV infection. Consequently, many host cell-virus interactions have been studied in vivo 3,4,5,6,7,8,9,10 . Although these studies are informative, they involve the use of infected birds which suffer from diseases that can be severe.
The CD40 ligand is a molecule that induces B cell proliferation 11 . Following identification of the gene encoding chicken CD40L (chCD40L), a soluble fusion protein was engineered that, when added to the culture media, induced the proliferation of chicken primary B cells ex vivo 12 . In 2015, B cells cultured in this fashion were found to support the replication of MDV 2 and in 2017, we and others found that primary bursal cells stimulated with chCD40L could be used as a model to study IBDV replication 13,14 . Here, we describe the isolation and culture of chicken primary bursal cells, the infection of the cells with IBDV, and the quantification of viral replication. Although we describe the method in the context of the BF, this could be applied to isolate and culture cells from different lymphoid organs.

Culture of Chicken Primary Bursal Cells
1. Centrifuge the cell suspension at 400 x g for 5 min and resuspend them in 1x complete IMDM. 2. Take an aliquot of the cell suspension, add it to a Trypan blue solution and count the number of viable cells that exclude the Trypan blue.
Determine the number of cells and the percentage viability. 3. Centrifuge the cell suspension at 400 x g for 5 min and resuspend it in complete IMDM supplemented with a 1:20 dilution of chicken CD40L at a density of 1 x 10 7 cells/mL. Titrate the concentrated supernatant containing the chCD40L to determine the optimal dilution, which is likely to lie in the range of 1:10 to 1:50. 4. Culture the cells in either 96-or 24-well plates at 37 °C for 48-72 h. U-bottomed 96-well plates are preferable to flat-bottomed plates.
NOTE: Cells can also be cultured at 40 °C

Chicken Primary Bursal Cells Can Be Cultured in the Presence of Chicken CD40L
When chicken primary bursal cells were cultured in the presence of soluble chCD40L, the number of cells increased fourfold from 9.02 x 10 5 to 3.63 x 10 6 per mL over a period of 6 days, in contrast to when it was absent (p <0.05) ( Figure 1A). Cell viability was also significantly improved, for example from 25% at day 3 post-culture in the absence of chCD40L to 48% in the presence of chCD40L (p <0.05) (Figure 1B) 13 .

Chicken Primary Bursal Cells Can Support the Replication of Both Cell-culture Adapted and Very Virulent Strains of IBDV
Mock-infected and infected cell cultures were fixed 18 h postinfection, labeled with a monoclonal antibody against IBDV VP2 and a secondary antibody conjugated to Alexa Fluor 488, and counterstained with DAPI. Infected cells had evidence of green fluorescence around the nucleus (Figure 2A), consistent with the presence of IBDV in the cytoplasm. This was evident for two strains of IBDV, a cell-culture adapted strain, D78, and a very virulent strain, UK661 (Figure 2A). At 5, 18, 24, and 48 h postinfection, RNA was extracted from infected cultures and subjected to RT-qPCR with primers specific to a conserved region of the IBDV VP4 gene. The expression of VP4 was first normalized to the house-keeping gene TBP and then expressed as fold change relative to mock samples in a ΔΔCt analysis. IBDV VP4 expression increased to 16,603 copies at 48 h postinfection with D78 and 38,632 copies at 48 h postinfection with UK661. Taken together, these data demonstrate that the chicken primary bursal cells could support the replication of cell-culture-adapted and very virulent IBDV strains

Discussion
In this study, we describe the successful culture of chicken primary bursal cells ex vivo in the presence of soluble chCD40L and demonstrate that these cells can support the replication of an attenuated strain and a very virulent strain of IBDV. This ex vivo model can be used to determine how the cells respond to an IBDV infection 13 , which has distinct advantages over in vivo and in vitro studies.
When harvesting the BF, it is critical to not puncture the gut so as to avoid bacterial contamination of the isolated bursal cells. In addition, it is important to isolate the primary cells as soon as possible after the organ harvest to limit cell death. The need to use chCD40L is a limitation of the technique; however, work conducted by Soubies et al. shows that the use of phorbol 12-myristate 13-acetate (PMA) to prolong bursal cell viability instead of chCD40L 14 may enable the model to be adopted by a greater number of laboratories. The protocol outlined above determines the optimal concentration of chCD40L empirically, by culturing primary B cells in serially diluted concentrations of the molecule and observing cell proliferation and viability. One potential modification to the protocol could be to purify the chCD40L molecule and to add a specific concentration to the cell culture media to avoid batch-to-batch variability.
In vivo studies have shown that following IBDV infection, there is an increase in the expression of genes involved in pro-inflammatory cytokine responses, Type I IFN responses, and apoptosis in the BF 5,9,10 . However, following infection, there is an influx of inflammatory cells and effector T cells into the BF, which differ in the genes they express compared to the infected B cell population 9 . It is, therefore, difficult to interpret how infected cells respond to IBDV. To address this, some research groups have characterized the transcriptional response of cells infected with IBDV in culture 16,17,18,19,20 . These in vitro studies have the advantage of well-defined MOIs and time-points postinfection. However, in vitro studies have typically been characterized in either fibroblast cells 16,17,20 or dendritic cells 18 . While providing some insight into host cell-IBDV interactions, the current belief is that the infection of B cells is crucial to the pathogenesis of IBDV and, therefore, the relevance of the data cannot be overinterpreted. Prior to our ex vivo bursal cell culture model, only one study had characterized the cellular response of B cells to IBDV infection 19 ; however, this study utilized an immortalized B cell line that was transformed due to infection with ALV, limiting the conclusions that could be made. In contrast, the ex vivo model of IBDV infection described here allows researchers to retain the advantages of in vitro studies, such as defined MOIs and time-points, while studying the interactions of the virus with its relevant host cell. As the primary bursal cells are obtained from uninfected BF tissue, there are no inflammatory or T cells present, and we have demonstrated by flow cytometry (using standard conditions) that, following chCD40L stimulation, 97% of the cell population is positive for the B cell marker Bu-1 (data not shown). Given that 3% of the cells are Bu-1 negative, it will be interesting to determine whether these cells become infected with IBDV and explore their gene expression and contribution to the pathogenesis.