Research Article: Humoral immune response to adenovirus induce tolerogenic bystander dendritic cells that promote generation of regulatory T cells

Date Published: August 20, 2018

Publisher: Public Library of Science

Author(s): Thi Thu Phuong Tran, Karsten Eichholz, Patrizia Amelio, Crystal Moyer, Glen R. Nemerow, Matthieu Perreau, Franck J. D. Mennechet, Eric J. Kremer, Patrick Hearing.


Following repeated encounters with adenoviruses most of us develop robust humoral and cellular immune responses that are thought to act together to combat ongoing and subsequent infections. Yet in spite of robust immune responses, adenoviruses establish subclinical persistent infections that can last for decades. While adenovirus persistence pose minimal risk in B-cell compromised individuals, if T-cell immunity is severely compromised reactivation of latent adenoviruses can be life threatening. This dichotomy led us to ask how anti-adenovirus antibodies influence adenovirus T-cell immunity. Using primary human blood cells, transcriptome and secretome profiling, and pharmacological, biochemical, genetic, molecular, and cell biological approaches, we initially found that healthy adults harbor adenovirus-specific regulatory T cells (Tregs). As peripherally induced Tregs are generated by tolerogenic dendritic cells (DCs), we then addressed how tolerogenic DCs could be created. Here, we demonstrate that DCs that take up immunoglobulin-complexed (IC)-adenoviruses create an environment that causes bystander DCs to become tolerogenic. These adenovirus antigen loaded tolerogenic DCs can drive naïve T cells to mature into adenovirus-specific Tregs. Our study reveals a mechanism by which an antiviral humoral responses could, counterintuitively, favor virus persistence.

Partial Text

Human adenoviruses (HAdVs), of which there may be 85 types (based on serology and genome analyses), typically cause self-limiting respiratory, ocular, and gastro-intestinal tract infections in immunocompetent individuals. After repeated encounters, most young adults generally harbor cross-reactive, long-lived humoral and T-cell responses [1–3] that are thought to work together to efficiently blunt subsequent HAdV-induced morbidity. However, in spite of the robust anti-HAdV immune responses, HAdVs routinely establish decades-long, subclinical infections that are characterized by low level shedding of progeny virions [4,5]. While potential molecular mechanisms by which HAdVs evade the immune response have been proposed [6], we suspected that complementary mechanisms also exist. Of note, in T-cell compromised patients the loss of cellular control of persistent HAdV infection can lead to fulminant and fatal disease [4,5]. It is noteworthy that serological evidence that the patient has been infected by a given HAdV type before hematopoietic stem cell transplantation is predictive of escape from the same HAdV type during immune suppression [7].

HAdV infections lead to multifaceted, robust, long-lived cellular and humoral responses in most young immunocompetent adults. Nonetheless, several HAdV types somehow circumvent immune surveillance to establish persistent infections. It is well documented that HAdV neutralizing antibodies are type specific, while the anti-HAdV cellular response is cross-reactive [1,3,8,47–49]. In addition, it is the anti-HAdV cellular response that protects us from reactivation of persistent infections. The dichotomy between the two arms of the adaptive immune response led us to address how anti-HAdV antibodies influence anti-HAdV T-cell responses. In this study, we initially asked if healthy adults harbor HAdV-specific Tregs, which would be indicative of a path towards HAdV persistence. We then explored how tolerogenic DCs and HAdV-specific Tregs could be generated. We previously showed that IC-HAdV5s are internalized by, and aggregate in, DCs [14]. Following protein VI-dependent endosomal escape of the capsid, the viral genome is engaged by AIM2 in the cytoplasm. AIM2 nucleation induces ASC (apoptosis-associated speck protein containing a caspase activation/recruitment domain) aggregation, inflammasome formation, caspase 1 auto-activation, pro-IL-1β and GSDMD cleavage, and GSDMD-mediated loss of cell membrane integrity. Here we demonstrate that the pyroptotic environment induced by IC-HAdV5 plays a significant role the creation of tolerogenic bystander DCs. We further show that some of these bystander DCs can induce HAdV-specific memory T cells to proliferate, and/or drive naïve CD4 cells towards a Treg phenotype. The Tregs generated in this ex vivo assay are capable of inhibiting the proliferation of anti-HAdV T cells. We therefore propose that an antiviral humoral responses can, counterintuitively, abet HAdV persistence.