Summary: Autoimmunity Modifying Biologicals
Autoimmunity Modifying Biologicals (AIM Biologicals) induce selective, antigen-specific immunological tolerance in an organ of choice via soluble, mostly physiological proteins that also protect embryos from the maternal immune system. Autoimmune diseases are driven by rare autoantigen-specific T cells that react towards autoantigens presented on classical MHC molecules and attack a healthy organ (1). Currently available non-selective immunosuppressants inhibit these autoreactive T cells, but also inhibit protective immune functions (2). Thus, doses high enough to completely stop progression of the autoimmune disease cannot be administered (3). At the same time, patients have an increased risk for severe opportunistic infections (4).
We discovered that the non-classical MHC Ib molecule HLA-G, which also exists in soluble forms, accumulates in lymph nodes and induces selective immune tolerance towards the presented peptide. Physiologically, this mechanism enables embryos to induce embryo-specific tolerance in maternal immune cells (1). AIM Biologicals use optimized HLA-G to induce targeted and effective tolerance towards autoantigens attacked during autoimmune diseases (2). These therapeutics should not affect protective immune responses that protect a patient from infections or tumors (3). Thus, AIM Biologicals can be used to stop autoimmune disease progression (4) without causing opportunistic infections (5).
Introduction
MHC Class Ib Molecules
Immunosuppressive MHC class Ib molecules such as HLA-G are critical for tolerance induction during pregnancy. They exert immunosuppressive effects on various immune cells via immunosuppressive receptors such as ILT2, ILT4 and Kir2DL4.
While classical MHC molecules present intracellular antigens at the cell surface, HLA-G also exists in soluble isoforms. We found that such soluble HLA-G molecules bind to the surface of antigen-presenting cells (primarily via ILT4/LILRB2). When attached to the surface of antigen-presenting cells, these soluble MHC molecules remain functional (1). This mechanism enables embryo- or tumor-derived HLA-G molecules to act in draining lymph nodes where they can potently influence the priming of host immune responses.
HLA-G & Antigen-Specific Immunosuppression
Our immune systems are made to attack all foreign tissues, including human ones, which often makes it difficult to find suitable organ donors. In stark contrast, foreign human antigens which come from the father are tolerated during pregnancy. And immunosuppression observed during pregnancy is very different from immunosuppression induced with currently available therapeutics: it is selective, it hardly causes relevant side effects. In fact, pregnancy even reduces relapse rates in patients with MOG antibody disease (MOGAD) or multiple sclerosis (MS). Thus, we wondered if there may be unknown mechanisms that induce antigen-specific tolerance during pregnancy. Although HLA-G was often described as a regular immunosuppressive checkpoint molecule, the presented peptides had been described as less relevant.
To test wether HLA-G could excert peptide-specific immunosuppressive effects, we challenged mixed, highly activated cytotoxic CD8+ T clones that recognize to two different peptides either with HLA-G expressing cells alone, or with HLA-G expressing cells loaded with one of the recognized peptides. Within 16 hours, more than 90 % of the cytotoxic T cells recognizing the HLA-G presented peptide (red) were gone or undergoing programmed cell death (apoptosis), while the non-targeted T cell clone (grey) remained largely unaffected. This mechanism could be essential to eliminate embryo-specific cytotoxic T cells duing pregnancy.
The AIM Biological Platform
AIM Biological generations
While terminally differentiated T cells become apoptotic when they encounter their antigen on HLA-G, naïve T cells are not prone to undergoing apoptosis. Also, physiological HLA-G consists of a peptide, the HLA-G domains and beta-2-microglobulin (b2m) and is a fairly unstable molecule. In order to improve stability and to further explore antigen-specific effects of peptide-loaded HLA-G also on naïve T cells or murine T cells, the HLA-G molecule was modified into a single chain molecule. We first designed soluble molecules comprising a peptide antigen, HLA-G and b2-microglobulin (b2m), and connected these three molecules via covalent linkers (2nd generation AIM Biologicals). 3rd generation AIM Biologicals may possess antigen-binding a1 and a2 domains of other MHC class I molecules to enhance the flexibility and versatility of these molecules. 3rd generation AIM Biologicals can also contain murine MHC molecules if needed for animal studies. Such, constructs comprising the a1 and a2 domains of murine H2-Kb can present the ovalbumin-derived peptide SIINFEKL to OT-1 T cells. (OT-1 T cells express a transgenic T cell receptor that specifically recognizes this antigen). Our 2017 platform patent claims the use of HLA-G in combination with self-antigens expressed in autoimmune affected organs.
Induction of antigen-specific, tolerogenic cells in vitro
We used Ovalbumin-specific AIM Biologicals and gp34-specific control AIM Biologicals to show that a cognate AIM Biologicals loaded on dendritic cells (DCs) suppresses the secretion of pro-inflammatory cytokines like IFN-g. The secretion of anti-inflammatory IL-10 in regulatory T cells (Treg) was induced. No such effect was observed with AIM Biologicals presenting a gp34 control peptide that is not recognized by OT-1 T cells. These in vitro ELISPOT results highlight the potential of AIM Biologicals to induce regulatory T cells in an antigen-dependent manner (left). Ovalbumin-specific AIM Biologicals also inhibit the killing of OVA-loaded target cells by OT-1 T cells (right).
AIM Biologicals selectivels induce Treg, prevent MS symptoms and autoantibodies in animal models
First animal studies indicate that surrogate AIM Biologicals selectively induce tolerogenic, IL-10 secreting T cells and prevent EAE symptoms and autoantibodies in vivo.
A) C57Bl/6J wild-type mice were treated with a single dose of an AIM Biological presenting either a viral peptide (AIM-gp34) or a model antigen (AIM-Ova). After 14 days, mice were sacrificed and splenocytes were challenged with gp34 (gpP) or Ovalbumin (ovP) peptide. Expansion or polarization of IL-10 secreting regulatory T cells (detected via ELIspot) only occured in mice that had been treated with the corresponding AIM Biological. B) C57Bl/6J ODC-Ova mice express the model antigen Ovalbumin in oligodendrocytes. After adoptive transfer of OT-I CD8+ T cells, these mice develop EAE symptoms. Here, 5 mice per group were treated with a single dose of an AIM Biological presenting either a viral control peptide (AIM-Gp34) or the target antigen (AIM-Ova). ** indicate P<0.01 significant rescue in AIM-Ova group vs effectors only (red) or control AIM-Gp34 (green) treatment (two-way Anova). C) MOG-tolerance inducing AIM Biologicals selectively prevent the emergence of MOG-specific autoantibodies in a MOG35-55 induced EAE model. Autoantibodies are harmful in many autoimmune diseases. Total IgG levels remained unaltered (not shown).
AIM Biologicals significantly reduce optic neuritis symptoms after disease onset
Further animal studies showed that surrogate AIM Biologicals that induce selective tolerance to a MOG peptide have excellent therapeutic effects in a strong genetic model, in which all T cells express the MOG-specific 2D2 T cell receptor. Ocular inflammation could be significantly reduced (Figure 6A), paralysis (EAE) was completely prevented (Figure 6B), and loss of neurons or oligodendrocytes through apoptosis (cleaved caspase 3) could be completely prevented in all assessed MOGAD typically affected organs like the eye (retina, C), optic nerve (D) or spinal cord (E). T cell infiltration and demyelination was also almost completely prevented (data not shown).
Human candidates successfully prioritized for MOGAD/MS and Type 1 Diabetes
In parallel to the animal studies, we have prioritized human candidate molecules for both myelin antigens (usable in MOG antibody disease and multiple sclerosis) and for Type I Diabetes (T1D) antigens.
candidate prioritization process and results
A) >30 candidates were prioritized based on bioinformatics tools and literature analyses and then cloned and expressed in transient transfection cell culture. Candidates with acceptable production yields are then prioritized with regards to Treg induction in (in this case 55) healthy donors and patient PBMCs (not shown, equally effective). B) Overview of prioritization results for type I diabetes candidates. Good candidates induce Treg in up to 80% of healthy donors PBMCs.
Conclusions
During pregnancy (when HLA-G is physiologically expressed), it is very likely that soluble HLA-G loaded with fetal antigens can selectively delete antigen-specific effector cells that could harm the embryo. Naïve T cells recognizing fetal tissue antigens are polarized towards a regulatory phenotype. Interestingly, this selective
tolerance induction against fetal, i.e. paternal and maternal antigens also provides a possible explanation for the commonly observed amelioration of autoimmune diseases during pregnancy. The AIM Biologicals platform employs these mechanisms to develop molecules that can induce antigen-specific tolerance in tissues targeted by autoreactive T cells. We have proven the efficacy of this approach in now 4 different animal models and successfully prioritized candidates for MOGAD and T1D. We are excited to move this unique approach forward in order to provide millions of patients with a therapeutic option that may completely stop disease progression and reduce existing symptoms without causing severe side effects.
Interested in learning more? Please contact us!