Type 1 Diabetes (3) Flashcards
Organ specific autoimmunity - Rogue immune cells target the pancreatic beta cells
From thymus to beta cell destruction: T1D pathogenesis
Thymus
>Developing T cells undergo selection
>positive selection to progress further and enter the periphery, or negative selection because they react to self-antigens and should be deleted (central tolerance)
>Those at risk of developing T1D have T cells that are self-reactive to beta cell antigens (failure of negative selection)
> organ specific autoimmune disease
represents a failure of self from non-self discrimination (basically a fundamental task of the immune system)
> Pancreatic lymph node
In the periphery, these naive t cells become activated, presumably in the draining lymph node or in the pancreatic islets themselves
due to the high conc of beta cell antigens present in both these places presented by the APCs that picked them up in the beta cells and moved back to the draining lymph nodes
activate the naive t cells that have escaped the thymus and these self-reactive t cells go on to infiltrate the islets and mediate the destruction of beta cells, resulting in dysregulation of glucose homeostasis
Human leukocyte antigen (HLA) locus on Chr6
T1D is an autoimmune disease involving the adaptive immune response
HLA Class II molecule:
>(alpha chain + beta chain)
>present peptides to CD4+ T cells
HLA Class I molecule:
>(alpha chain + beta-2-microglobulin chain (chr15))
>present peptides to CD8+ T cells
T cells and function
CD8+ T cells recognise peptides presented by HLA Class I molecules >cytotoxic function >directly kill the cells presenting that peptide on the HLA class I molecule
CD4+ T cells recognise peptides presented by HLA class II molecules >usually from a professional APC >will activate the APC to go on to do other functions e.g might be a B cell which goes on to produce (in the case of T1D) autoantibodies >helper function
Alleles for genes encoding MHC molecules confer the highest risk of T1D
HLA/MHC Class II (DQ locus)
>DQ2 - OR>3.6
>DQ8 - OR>11.4
>DQ2/DQ8 - OR>16.6
DQ locus consists of 2 genes
>gene that encodes the beta chain
>gene that encodes the alpha chain
>DQ2 and DQ8 is shorthand nomenclature to tell you what those alleles are for those 2 genes
HLA class II alleles confer the highest risk of type 1 diabetes
HLA DR3_DQ2, DR4-DQ8 heterozygosity confers the highest risk of T1D
OR>16.6
High risk alleles can present beta-cell antigen-derived peptides to T cells
HLA Class II molecules
>shape the T-cell repertoire during T-cell development
>Activate CD4+ T cells in the periphery
HLA locus that encodes these MHC genes is often highly associated with various autoimmune diseases
Why is the HLA class II region associated so closely with T1D (and other autoimmune diseases)?
Ongoing hypothesis:
- Beta-cell antigens may not be presented efficiently in the thymus by particular HLA molecules (e.g. DQ2 or DQ8)
- This enables the escape of low-affinity T cells from the thymus and subsequent activation by beta cell antigens that are present at high concentration in the pancreas and draining lymph nodes
(at even higher levels in the draining lymph nodes and the islets where the beta cells reside)
(if affinity is low, then the high conc of the b cell antigen at these places will help the autoantibody binding)
Genetic studies implicate non-HLA genes in T1D
Most of the alleles for these gene regions contribute relatively small risk <2.5 odds ratio
2 genes that have alleles with highest odds ratio that are non-HLA genes
>INS
>PTPN22
> > many of these genes are associated with immune function
Genetic polymorphisms for insulin associated with thymic expression and T1D risk
Key concept:
Dysregulation of negative selection generates a peripheral pool of anti-self-T cells with enhanced pathogenic potential
(displaying increased affinity and avidity)
A lot of self antigens are expressed in the thymus and presented by various cells in the medullary, so when the T cells interact with them, it sends out a signal to them to be deleted before they can get out into the periphery
> > INS = insulin
key autoantigen for human T1D
>VNTR upstream element of INS (variable number of tandem repeats)
»>Class I (26-63 repeats) = predisposing allele
»>Class III (140-210 repeats) = protective allele
> > T cell respnses
Genetic polymorphisms that affect thymic insulin expression are associated with T1D
Genetic polymorphisms that affect thymic insulin expression are associated with T1D
> > > VNTR upstream element of INS (variable number of tandem repeats)
»Class I (26-63 repeats) = predisposing allele
»Class III (140-210 repeats) = protective allele
> decreased thymic expression (of INS gene) = increased T1D risk
(T cells that recognise insulin not going to be negatively selected)
> increased thymic expression = decreased T1D risk
Reduced thymic insulin expression is predicted to:
> reduce negative selection of insulin-specific single positive thymocytes (precursor to the developed t cell that leaves and enters periphery)
> limit thymic development of beta cell-specific regulatory T cells (FOXP3+ CD4+ Tcells)
An allele for PTPN22 is associated with defective negative selection and T1D risk
Key concept:
Dysregulation of negative selection generates a peripheral pool of anti-self-T cells with enhanced pathogenic potential
(displaying increased affinity and avidity)
PTPN22 - Protein tyrosine phosphatase non-receptor 22
(is expressed in developing thymosites that become t cells in the thymus)
>genetic variant (R620W, 1858C>T mutation) associated with increased phosphatase activity and increased T1D risk
>PTPN22 is a negative regulator of T cell receptor (TCR) signalling
PTPN22 is a negative regulator of T cell receptor (TCR) signalling
Inhibiting various components of the intracellular signalling pathway, limiting transcription factors like NFAT and AP-1 binding to the DNA and causing expression of other genes
Elevated phosphatase activity (PTPN22) is predicted to:
Reduce TCR signalling and diminish apoptosis induction in beta cell-specific thymocytes
>autoreactive T cells escape to the periphery
> similar to insulin, limits thymic development of beta cell-specific regulatory T cells (FOXP3+ CD4+ T cells)
Recap on PTPN22 and INS association with defective negative selection and TD1 risk
During negative selection
>strong signal to developing thymocytes
>cause them to undergo apoptosis and die
> when you have PTPN22 allele, get increased phosphatase activity
reduce TCR signalling
reduces apoptosis induction in thymocytes that recognise B cell antigen and escape into periphery
> When you have reduced insulin expression in thymus
thymocytes who have low affintty for insulin can escape
once they get to pancreas lymph nodes, high amount of insulin makes up for that low affinity
cause immune response
PTPN22 is a negative regulator of B cell receptor (BCR) signalling
Elevated phosphatase activity is predicted to
>reduce BCR signalling and diminish apoptosis induction in autoreactive B cells
Beta-cell antigen-specific autoantibodies detected in T1D patients
If there are genetic defects affecting central tolerance, there should be evidence of autoreactive B cells and T cells in the periphery that have escaped negative selection
A number of the islet-cell specific autoantibodies (ICAs) that bind to components of the islets of Langerhans have been identified
Present in serum of people with T1D
Islet self-antigens
>insulin (produced by pancreatic beta cells)
>GAD65 (glutamic acid decarboxylase, expression not exclusive to pancreatic beta cells)
>IA-2 (tyrosine phosphatase-like protein islet antigen 2)
(not exclusive to pancreatic beta cells)
>ZnT8 (zinc transporter 8)
(zinc plays a key role in storage and secretion of insulin (hexomer), highly expressed in the endocrine but absent in exocrine pancreas. Also detected in extra-pancreatic sites)
These are antigens expressed by the beta cells
Beta cell antigens
Insulin granule
>insulin
>IA-2 (islet antigen 2)
>ZnT8 (zinc transporter 8)
Not in insulin granule
>GAD65 (glutamate decarboxylase)
Detection of islet-specific autoantibodies
Radiobinding assay
>to detect islet-specific autoantibodies (ICAs)
1) Collect serum from patient with suspected T1D
>serum containing autoantibodies
+
>radio-labelled insulin (or other beta-cell antigen)
2) Incubate and allow antigen/antibody complexes to precipitate
3) Measure radioactivity in the precipitate
>the amount of insulin specific antibody is proportional to the amount of radioactivity in the precipitate
>antibodies will start to complex and form these lattices which will precipitate in solution
Note:
Serum will contain a large collection of antibodies, with different specificities, only some will be ICAs
Association of islet autoantibodies and clinical presentation:
Positive seroconversion to islet autoantibodies in at-risk individuals
Positive seroconversion to islet autoantibodies in at-risk individuals
>rare before 6 months of age
>majority who develop T1D will be islet autoantibody positive by 3 years of age
(prospective studies - look at people at risk and wait to see if they develop autoantibodies, becomes an interesting biomarker to predict who will and wont develop T1D)
Association of islet autoantibodies and clinical presentation:
Not all islet autoantibodies are the same
Not all islet autoantibodies are the same
>Insulin autoantibodies (IAAs) are typically the first to be detected
>Glutamic Acid Decarboxylase Autoantibodies (GADAs) (GAD65) and IAAs are most frequent in childhood
> high affinity IAAs and GADAs associated with progression to multiple islet autoantibodies and diabetes
other islet autoantibodies (e.g. IA-2, ZnT8) typically appear later and indicate further progression of disease
> high affinity IAAs and GADAs associated with progression to multiple islet autoantibodies and diabetes
other islet autoantibodies (e.g. IA-2, ZnT8) typically appear later and indicate further progression of disease
Note:
During the progression as you havent yet developed clinical type 1 diabetes, you can start to see changes in these autoantibodies
> affinity maturation (higher affinity against both insulin and GAD) and epitope spreading (autoantibodies against other different islet/beta cell antigens - ZnT8 and IA-2)
Are beta-cell antigen-specific autoantibodies directly pathogenic?
No.
>autoantibodies are not necessary to develop T1D
>doesnt mean that they might not be contributing in some fashion, but they are not critical if you have all other components in place to develop the disease
Islet autoantibodies do not have a direct cytotoxic effect on human islets in vitro
Beta cells dont just die because the antibodies are there in the same in vitro test tube
Transfer of maternal autoantibodies to the foetus does not increase the offspring’s risk of developing T1D
Not transferring disease in any fashion that we can pick up
A single case of a 14 year old boy with x-linked agammaglobulinaemia developed T1D
Primary immunodeficiency where you dont develop autoantibodies
Are beta-cell antigen-specific autoantibodies directly pathogenic?
Summary:
Remember:
B cells act as antigen presenting cells
B cell Receptors (i.e. membrane bound versionms of the secreted antibodies) capture beta cell autoantigens
>process and present peptides by HLA class II molecules to CD4+ T cells
>CD4+ T cells are activated and secrete cytokines that stimulate B cells to secrete antibodies
Beta-cell specific T cells are detected in the periphery of T1D patients
If there are genetic defects affecting central tolerance, there should be evidence of autoreactive B cells and T cells in the periphery that have escaped negative selection
Beta-cell-specific T cells
Rare in peripheral blood but can be detected in the blood of healthy individuals
> tells us that the thymus is not perfect in selecting/deleting T cells and we have peripheral tolerance (other systems in place) so even healthy individuals can have these self-reactive T cells
The phenotype of circulating beta-cell specific T cells is distinct in T1D patients vs healthy controls
> T cells from patients = effector/memory phenotype
(i.e. activated)
T cells from healthy controls = naive phenotype
> T cells from patients = proinflammatory cytokines
(e.g. IFNgamma)
T cells from healthy controls = regulatory responses (e.g. IL-10)
Different assays can be used to detect beta-cell specific T cells
T-cell proliferation assay: 5,6-carboxyfluorescein diacetate succinimidyl ester (CFSE)
T-cell proliferation assay: 5,6-carboxyfluorescein diacetate succinimidyl ester (CFSE)
> Peripheral blood mononuclear cells are stained with CFSE and incubated with different beta-cell antigens/peptides
(in PBMC population, you have both T cells and APCs so you mix everything together, see which cells are proliferating)
> Flow cytometry analysis indicates if cells have proliferated
>stain cells for CD4 or CD8 T-cell markers
>T cells that have reduced CFSE have proiferated (i.e. each time a cell divides, it loses half of its CFSE to each daughter cell)
Different assays can be used to detect beta-cell specific T cells
T-cell detection: MHC/peptide tetramer assay
T-cell detection: MHC/peptide tetramer assay
>fools the T cells into thinking its binding to a cell
>MHC is a class I with Beta-2-microglobulin, presenting a peptide
>link 4 of them to a biostrip-aviudin complex that has a fluorophore
>use Flow Cyto Fluorescence-activated cell-sorting (FACS) to identify which T cells bind to this MHC/peptide tetramer
Binding results in fluorescence
What are T-cell antigen targets associated with T1D?
Note: some overlap between B-cell autoreactive responses and T cell autoreactive responses
Peptide epitopes have been identified from several islet cell proteins (some also targeted by autoantibodies)
> Proinsulin (before c-peptide is cleaved)
GAD65 (not exclusive to pancreatic beta cells)
IA-2 (not exclusive to pancreatic beta cells)
ZnT8 (also detected in extra-pancreatic sites)
> IGRP (islet-specific glucose-6-phosphatase catalytic subunit-related protein)
IAAP (islet-amyloid polypeptide; co-secreted with insulin from beta cells)
Chromogranin A (beta cell secretory granule protein, also expressed in neuroendocrine tissue)
Insulin C-peptide specific CD4+ T cells are detectable in peripheral blood
T-cell proliferation CFSE assay
>insulin c-peptide responses detectable in peripheral blood of >60% of people with recent-onset T1D
(taking peptides from the c-peptide region and incubating with these APCs, looking at CSFE proliferation)
> > can see that number of the recent onset patients had T cells in the peripheral blood that recognize C-peptide indicating that they had autoreactive responses against the beta cell antigens
people with longstanding T1D dont have this because majority of beta cells gone, T cells quietened down, nothing to respond to so not much Beta cell autoreactive T cells in the periphery
What is happening in the islets?
Insulitis is the pathologic hallmark of T1D
>immune cell infiltration of the islets associated with beta cell loss
Differences in insulin content between islets in the same pacreas of T1D donor
You dont see every islet in a PT with T1D that shows very little or no insulin
> you see patches across the different sections
> pancreas quite big so you cant really take secretions all across and stain everything
taking snapshots to see what the insullitis (insulin production) and what the insulin staining is
Insulitis is often variable in pancreatic tissue from T1D patients
Insulitis is defined by at least 16 CD45+ cells (ie. immune cells) per islet present in 3 or more islets
Peri-insullitis = peripheral insulitis
Characteristics of insulitis in pancreatic tissue from T1D patients
> Only 10-30% of islets show insulitis at any time, even when tissue is obtained at diagnosis
> insulitis typically observed as peri-insulitis or intra-insulitis (immune cells in middle of islet) in insulin-positive islets
> Studies using various samples from various sources indicate
>infiltration is not homogeneous
>some part of the pancreas have more infiltration than others (i.e. insulitis is lobular)
>There are ‘pseudoatrophic’ islets devoid of beta cells
(islets devoid of beta cells but can see staining of glucagon and somatostatin producing cells)
Immune cells can be detected within and around the islets
With staining for CD8 and CD4 T cell markers, and CD68 macrophage marker
Immune cells can be detected within and around the islets
Immunofluorescence staining of serial sections of pancreatic tissue
> > additional staining of immune cells (CD45)
Various T cell, B cell, and Macrophage markers
Shows that APCs (macrophages, dendritic cells, B cells) are present within the islets
>example of peri-insulitis as insulin-producing beta cells are still present
How to clone and screen human islet-infiltrating T cells
i.e. how to make sure that the T cells in the islets are actually recognising Beta cell antigens and not there for other reasons
1) Grow out cells from donor T1D pancreas
2) Clone individual cells
3) Expand cells in culture
4) incubate T cells against a panel of synthetic peptides derived from beta-cell autoantigens
(looking for stimulation of these T cells which will produce cytokines in response)
5) measure T-cell activation based on cytokine secretion
Beta-cell specific CD4+ T cells can be detected in islets of T1D patients
Made smaller c-peptides and fine-tune the actual peptide that was presented by the APC to the T cell
>screening for T cell activation based on cytokine secretion
> Can see pro-insulin peptides resulted in secretion of IFNgamma
> Key finding
>26% of the CD4+ t cell clones recognised peptide epitopes derived fro the C-peptide of insulin
Beta-cell specific CD8+ T cells can be detected in islets of T1D patients
Pancreases from 45 T1D donors
>analysed using HLA Class I/peptide tetramers (FACS)
>tetrameers presented peptides from differnt beta-cell antigens: Insuin, IA-2, GAD65, IGRP
> Tetramer staining of pancreatic islets
> Key finding
both single and multiple CD8+ T cell autoreactivities against islet autoantigens (i.e. beta-cell proteins) were detected within islets of T1D patients, even up to 8 years after clinical diagnosis
More characteristics of insulitis in pancreatic tissue from T1D patients
> Both CD4+ and Cd8+ T cells are detected in the islets that recognise beta-cell-derived antigens
>these autoreactive T cells secreted IFNgamma and other inflammatory cytokines = effector T cells
Caveat: Histological studies examined only relatively small proportion of each pancreas
>would be nice to be able to histologically stain the whole pancreas to see how much insulitis or inflammation we can see throughout the whole pancreas
Detection of low-grade enteroviral infection in the islets of person with newly diagnosed T1D
Using an antibody against viral protein, able to detect what appears to be viruses within the islet
> Triggers such as viral infection and innate immune activation lead to type 1 interferon production
>increased IFNalpha is detected in pancreas of type 1 diabetic subjects
>interferon stimulated genes are overexpressed in pateints recently diagnosed with T1D and in islets of young NOD mice (animal model of T1D)
HLA Class I and II molecules are expressed in the islets
T1D > hyperexpression of Class I
Class II also detected in and around islet capillaries
> > Upregulation of HLA class I correlates with number of CD8+ T cells in the islets
Upregulation of HLA class I correlates with number of CD8+ T cells in the islets
Increased expression of HLA molecules and presentation of Beta cell autoantigens make beta cells more susceptible to destruction by infiltrating autoreactive T-cells
What do we think is happening? The suspected path of an autoreactive T cell
1) Naive Beta-cell autoreactive T cell escapes thymus
2) Activation and expansion of autoreactive T cell in pancreatic lymph node
>upregulation of activation markers
3) Beta-cell autoreactive T cells gain effector function and expand
>produce cytokines and cytotoxic molecules
>activated T cells may also leave the islets and recirculate in the blood
Autoreactive CD8+ T cells recognise and kill the beta cells
CD8+ T cells interact with Beta cells through TCR-HLA Class I
cytotoxic killing
Secrete perforin and granzymes which directly kills beta cells by inducing cell death
