Introducing tolerance to foreign and partly foreign proteins by erythrocyte binding and artificial glycosylations

Over the last few years, protein drugs have steadily gained importance in the clinic. Compared to small molecule drugs, protein drugs are taken up by antigen presenting cells, which degrade the proteins into peptides that are then presented on major histocompatability complex I and II to CD8+ respectively CD4+ T-cells. CD4+ T-cells are then able to activate B-cells, which upon activation will produce anti-drug antibodies. While anti-drug antibodies have mostly been viewed as a safety risk leading to drug injection related side effects, they have recently been proven to diminish the efficacy of the treatment. High antibody titers against Humira, a clinically approved monoclonal antibody used to treat rheumatoid arthritis, have been associated with low serum drug concentration and high disease activity score, ultimately leading to treatment failure. Two approaches to induce tolerance were developed in the Hubbell lab. One approach utilizes protein variants that are designed to bind to circulating erythrocytes. As 1 % of erythrocytes undergo apoptosis each day and are taken up by antigen presenting cells, along with the bound protein, they are presented to the immune system in a non-inflammatory tolerogenic fashion. Immunological tolerance is thereby induced against the apoptotic erythrocyte debris and the associated bound protein. The second approach uses synthetic glucose, galactose or mannose polymers conjugated to the target protein. The polymers target C-type lectins, which are known to clear apoptotic debris, leading to presentation of the protein by hepatic antigen presenting cells in a tolerogenic liver microenvironment. This thesis investigates the potential of these approaches to induce tolerance towards protein drugs. Arylsulfatase B was used as a model protein to investigate the potential of tolerance induction by targeting circulating erythrocytes, as 97% of patients receiving arylsulfatase B develop anti-drug antibodies. An erythrocyte binding variant of arylsufatase B was created by chemically conjugating the erythrocyte binding peptide ERY1 to arylsulfatse B. In mice receiving two doses of ERY1-arylsulfatase B one week apart followed by weekly doses of arylsulfatase B a significant delay of four week in the production of anti-drug Antibodies was found when compared to mice receiving weekly doses of only arylsulfatase B. To investigate the ability of synthetic glyco-polymers to induce tolerance towards protein drugs we used asparaginase as model protein. Between 37.5 % to 75% of patients treated with asparaginase develop anti-drug antibodies. Administering three intravenous injections of asparaginase-conjugated to glucose polymers before eight weeks of treatment with asparaginase significantly reduced anti-drug antibody titers by a factor of 125 when compared to animals only receiving eight weeks of treatment with asparaginase. Three pre-injections with asparaginase conjugated to the galactose or mannose polymer led to a reduction in anti-drug antibody titers by a factor of 12 respectively 400 after 5 weeks of treatment with asparaginase when compared to mice receiving only asparaginase for 5 weeks. These results indicates that synthetic glyco polymers, especially mannose and glucose based ones, are able to induce long term tolerance towards asparaginase. Synthetic glucose and mannose polymers therefore represent an excellent molecular approach to induce tolerance towards protein drugs.

New immunomodulatory roles of lymphatic endothelium and implications for immunotherapy

Efthymia Vokali

The lymphatic system serves a critical role in fluid homeostasis, lipid metabolism and immune surveillance. The growing appreciation of its implication in various diseases challenges the conventional view of lymphatics as a passive transport system. Traditionally, the lymphatic endothelium has been perceived as a structural scaffold with certain immunological functions but no active involvement in immunomodulation. In the lymph nodes (LNs), the main sites of immune regulation in the periphery, lymphatic endothelial cells (LECs) come to close contact with immune cells, suggesting potential interactions. Indeed, LECs have been recently shown to suppress dendritic cell maturation and present peripheral tissue and tumor antigen for CD8+ T cell deletion. While LECs have only begun to be acknowledged as active regulators of immunity, their function and relative contribution in shaping immune responses is as yet poorly understood. This thesis aimed to elucidate the direct role of LECs in the induction of CD8+ T cell immunity and tolerance. First, we demonstrated that murine LECs can actively scavenge and cross-present exogenous antigen to cognate CD8+ T cells under non-inflamed conditions. By utilizing an in vitro coculture system and the model antigen ovalbumin, we investigated the antigen-specific interactions between LECs and CD8+ T cells. LEC-educated CD8+ T cells proliferated, exhibiting an activated phenotype, however, they displayed early-generation apoptosis and failed to produce effector cytokines. Our findings establish LECs as antigen-presenting cells and suggest that they may assist in the maintenance of peripheral tolerance during homeostasis. The particular differentiation state of LEC-educated CD8+ T cells prompted us to investigate whether they are terminally tolerized or they could escape the dysfunctional state. We demonstrated that LEC-educated CD8+ T cells adopted a distinct phenotype with central memory-like characteristics and shared multiple functional properties with memory cells. Upon antigen re-encounter, LEC-educated CD8+ T cells mounted proliferative responses and generated cytotoxic effector cells. More importantly, they participated in anti-infectious immunity while preserving a secondary-memory persistent population. Our findings reveal a unique differentiation state of antigen-experienced CD8+ T cells, generated under steady-state conditions, which remain inactive but can be functionally reactivated upon antigenic inflammatory challenge. This previously unanticipated feature of LECs triggered questions for their antigen-presenting function in an inflammatory setting. We asked whether the previously observed extensive proliferation of LECs in the LN during inflammation might directly influence the induction of immunity. We employed an anti-VEGFR3 blocking antibody to inhibit LEC proliferation and thus, reduce the number of LECs following vaccine immunization. Alternatively, we generated the Prox1-Cre-DTR mouse model, allowing for specific ablation of LECs following administration of diphtheria toxin in vivo. Our findings advance our perception of the relative contribution of LECs in the establishment of adaptive immunity. This thesis elucidates the multifaceted immunological role of LECs and strongly suggests the importance of harnessing their immunomodulatory function to enhance current vaccines and immunotherapeutic strategies. Looking forward, our work will contribute to future advances in the clinic.

EPFL2016

Exploiting immunofunctional nanoparticles as vaccines and synthetic pathogens

Sai Reddy

The emergence of recombinant DNA technology has shifted research towards developing subunit vaccines, which use protein or peptide antigens instead of live/attenuated pathogens. Recent research has also shown that the best way to get sustained immunity is to deliver an antigen directly to antigen-presenting cells (APCs), specifically dendritic cells (DCs). However there are two difficulties to overcome in targeting DCs for use in subunit vaccines: (1) DCs are present in a low concentration in peripheral tissue (where typical vaccines are delivered) compared to secondary lymphoid organs such as lymph nodes. Thus delivering antigen to DCs in lymph nodes offers a powerful alternative for vaccine technology. (2) Activating DCs requires co-delivering an adjuvant also sometimes known as a molecular "danger signal" which matures DCs and initiates the subsequent adaptive immune response. Adjuvants are needed since most protein and peptide antigens are insufficiently immunogenic. Current danger signals being explored in experimental vaccine formulations mimic bacterial components (e.g., lipopolysaccharide (LPS), unmethylated CpG DNA) and often signal through a class of pattern recognition receptors knows as Toll-like receptors (TLRs), however limitations exist in that these often cause toxic side effects by inducing high levels pro-inflammatory cytokines. To address the need to target DCs, we used a nanoparticle delivery approach. We investigated the delivery of 20, 45, and 100 nm diameter poly(ethylene glycol)-stabilized poly(propylene sulfide) (PPS) nanoparticles to DCs in the lymph nodes. These nanoparticles consisted of a cross-linked rubbery core of PPS surrounded by a hydrophilic corona of poly(ethylene glycol). The PPS domain is capable of carrying hydrophobic drugs and degrades within oxidative environments. 20 nm particles were most readily taken up into lymphatics following interstitial injection, while both 20 and 45 nm nanoparticles showed significant retention in lymph nodes, displaying a consistent and strong presence at 24, 72, 96 and 120 h post-injection. Nanoparticles were internalized by up to 40-50% of lymph node DCs (and APCs) without the use of a targeting ligand, and the site of internalization was in the lymph nodes rather than at the injection site. Finally, an increase in nanoparticle-containing DCs (and other APCs) was seen at 96 h vs. 24 h, suggesting an infiltration of these cells to lymph nodes. Thus, PPS nanoparticles of 20-45 nm showed potential for vaccine delivery specifically targeting DCs in lymph nodes. Next we exploited our nanoparticle technology as a vaccine platform. We engineered antigen-bearing nanoparticle vaccines with two novel features: lymph node-targeting and in situ complement activation. Following intradermal injection, ultra-small nanoparticles (25 nm) were readily transported by interstitial flows into lymphatic capillaries, which leads to their accumulation in lymph nodes resident DCs. We further designed the nanoparticle surface chemistry to activate the complement cascade, thus spontaneously generating a danger signal in situ for DC activation. With ovalbumin as a model antigen conjugated to the nanoparticles, we demonstrated humoral and cellular immunity in mice. Finally we exploited nanoparticles as "synthetic pathogens" to isolate the effect of free surface thiols on complement activation and the subsequent immunological response induced. We found that surfaces with both hydroxyl and thiol groups activated complement to significantly higher levels than hydroxl groups alone. We then demonstrated that nanoparticles that induced thiol-associated complement activation were potent danger signals that were able to mature dendritic cells to express high levels of costimulatory molecules and the Th1 promoting cytokine interuleukin-12. Moreover, this dendritic cell maturation was dependent on a Toll-like receptor 4 (TLR4) signaling pathway and independent of the adaptor protein myeloid differentiation protein 88 (MyD88). Thus we uncover a potential direct link between complement and TLR4 signaling that is dependent on the specific activation of complement by thiol-expressing surfaces. In conclusion, this thesis develops a novel vaccine methodology, which is at the interface of nanotechnology and biotechnology, by harnessing lymphatic transport to target DCs and complement activation to activate adaptive immunity; we demonstrated an elegant vaccine technology platform for applications that include global health. Finally we exploit our nanoparticles as synthetic pathogens to discover fundamental mechanisms connecting the complement system and TLR signaling.

EPFL2008

Modes of antigen presentation by lymph node stromal cells and their immunological implications

Sachiko Hirosue

Antigen presentation is no longer the exclusive domain of cells of hematopoietic origin. Recent works have demonstrated that lymph node stromal cell (LNSC) populations, such as fibroblastic reticular cells, lymphatic and blood endothelial cells, not only provide a scaffold for lymphocyte interactions but also exhibit active immunomodulatory roles that are critical to mounting and resolving effective immune responses. Importantly, LNSCs possess the ability to present antigens and establish antigenspecific interactions with T cells. One example is the expression of peripheral tissue antigens, which are presented on major histocompatibility complex (MHC)-I molecules with tolerogenic consequences on T cells. Additionally, exogenous antigens, including self and tumor antigens, can be processed and presented on MHC-I complexes, which result in dysfunctional activation of antigen-specific CD8+ T cells. While MHCI is widely expressed on cells of both hematopoietic and non-hematopoietic origins, antigen presentation via MHC-II is more precisely regulated. Nevertheless, LNSCs are capable of endogenously expressing, or alternatively, acquiring MHC-II molecules. Transfer of antigen between LNSC and dendritic cells in both directions has been recently suggested to promote tolerogenic roles of LNSCs on the CD4+ T cell compartment. Thus, antigen presentation by LNSCs is thought to be a mechanism that promotes the maintenance of peripheral tolerance as well as generates a pool of diverse antigen-experienced T cells for protective immunity. This review aims to integrate the current and emerging literature to highlight the importance of LNSCs in immune responses, and emphasize their role in antigen trafficking, retention, and presentation.