Activation of Langerhans-Type Dendritic Cells Alters Human Cytomegalovirus Infection and Reactivation in a Stimulus-Dependent Manner

Abstract

Oral mucosal Langerhans cells (LC) are likely to play important roles in host defense against infection by human cytomegalovirus (CMV). We previously showed that in vitro-differentiated immature LC (iLC) populations contain smaller amounts of infected cells but produce higher yields than mature LC (mLC) cultures, obtained by iLC stimulation with fetal bovine serum (FBS), CD40 ligand (CD40L) and lipopolysaccharide (LPS). Here, we sought to determine if exposure to select stimuli can improve LC permissiveness to infection, if specific components of the mLC cocktail are responsible for lowering viral yields, if this is due to defects in progeny production or release, and if these restrictions are also effective against reactivated virus. None of the stimuli tested extended the proportion of infected cells to 100%, suggesting that the block to infection onset cannot be fully removed. While CD40L and FBS exerted positive effects on viral progeny production per cell, stimulation with LPS alone or in combination with CD40L was detrimental. Reductions in viral titers were not due to defects in progeny release, and the permissive or restrictive intracellular environment established upon exposure to each stimulus appeared to act in a somewhat similar way toward lytic and latent infections.

Single-cell transcriptome analysis of CD34+stem cell-derived myeloid cells identifies a CFU-GEMM-like population permissive to human cytomegalovirus infection

“…Differentiated cells were then activated by exposure to GM-CSF, FBS, CD40L and LPS, and infected with CMV strain TB40/E. Consistent with our previously published data [14–16], cell numbers did not increase over time (not shown), and only 3 ± 1.5 % of non-activated but 10 ± 5 % of activated cells expressed the viral IE1/IE2 proteins at day two post-infection (pi), notwithstanding the use of a multiplicity of infection (MOI) of ten pfu/cell, which is sufficient to infect the totality of permissive cell types such as fibroblasts (Fig 1A and B). Despite containing higher numbers of IE1/IE2 + cells at each time point, activated cell populations produced lower intracellular progeny amounts per IE1/IE2 + cell than non-activated cells (Fig 1C and D).…”

“…In previous studies, we showed that activated and non-activated myeloid cells differentiated from CD34 + HSC are semi-permissive to CMV infection [9, 14–16]. As such, they constitute a useful model to identify the cellular determinants of viral tropism.…”

“…Cell staining was performed as previously described [16]. Briefly, cytospin preparations of myeloid cells were fixed in 1.5% formaldehyde for 30 min, permeabilized in 0.5% Triton-X 100 for 20 min, and blocked in 40% FBS/40% goat serum for 30 min before incubation with antibodies directed against the viral proteins IE1/IE2 (MAb810, 1:600, or AF488 MAB810X, 1:200, Millipore, Temecula, CA), UL32 (pp150, 1:400, a kind gift from Bill Britt, University of Alabama, Birmingham), UL84 (1:500, Virusys, Taneytown, MD), UL44 (1:200, Virusys, Taneytown, MD), or UL57 (1:100, Virusys, Taneytown, MD) for one hour, followed by secondary antibodies conjugated to Alexa-Fluor 488 or Alexa-Fluor 594 (1:200, Invitrogen, Carlsbad, CA, and Jackson Immunoresearch, West Grove, PA) for another hour.…”

“…We previously showed that exposure of cord blood CD34 + HSC to specific cytokines such as Flt3 ligand (FL) and transforming growth factor β1, known to instruct their differentiation into Langerhans cells [12, 13], gives rise to myeloid cell populations that are semi-permissive to CMV infection. While only 2-3% of non-activated cells obtained at the end of the differentiation period allowed expression of the viral immediate early 1 and 2 (IE1/IE2) proteins, which are essential for infection onset and progression, cell activation by exposure to granulocyte-macrophage colony-stimulating factor (GM-CSF), fetal bovine serum (FBS), CD40 ligand (CD40L) and lipopolysaccharide (LPS) partially released this initial block, raising the proportion of IE1/IE2 + cells by 5-10 fold [14–16]. Unexpectedly, however, non-activated cells produced higher yields per IE1/IE2 + cell than activated cells, suggesting that signaling by GM-CSF, FBS, CD40L and LPS may trigger the establishment of a second block to infection progress, acting after IE gene expression and negatively impacting viral progeny production.…”

“…Unexpectedly, however, non-activated cells produced higher yields per IE1/IE2 + cell than activated cells, suggesting that signaling by GM-CSF, FBS, CD40L and LPS may trigger the establishment of a second block to infection progress, acting after IE gene expression and negatively impacting viral progeny production. This second block is unlikely to be due to defects in progeny release, as non-activated and activated cells generated similar ratios of cell-free to cell-associated virus [16], but may instead depend on impairments in the assembly of viral replication compartments (Galinato and Hertel, unpublished).…”