model diseases for studies of autoimmunity
TRANSCRIPT
ACTAUNIVERSITATIS
UPSALIENSISUPPSALA
2015
Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1154
Model diseases for studies ofautoimmunity.
FRIDA DALIN
ISSN 1651-6206ISBN 978-91-554-9392-9urn:nbn:se:uu:diva-265276
Dissertation presented at Uppsala University to be publicly examined in Aula Gunnesalen,Akademiska sjukhuset ingång 10, 751 85, Uppsala, Wednesday, 16 December 2015 at09:15 for the degree of Doctor of Philosophy (Faculty of Medicine). The examination willbe conducted in Swedish. Faculty examiner: Professor Lucia Mincheva-Nilsson (UmeåUniversity, Department of Clinical Immunology).
AbstractDalin, F. 2015. Model diseases for studies of autoimmunity.. Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty of Medicine 1154. 44 pp. Uppsala: ActaUniversitatis Upsaliensis. ISBN 978-91-554-9392-9.
The events triggering autoimmune diseases are to large extent unknown and model diseasesare an important tool in studies aiming to elucidate molecular mechanisms in autoimmunity.Autoimmune Addison’s disease (AAD) is a rare disease characterized by autoimmunedestruction of adrenal glands and most patients with AAD have autoantibodies against theenzyme 21#hydroxylase in the adrenal cortex. The autoimmune destruction in AAD is howeversuspected to be initiated by T cells. One of the most important investigations in this thesis wasto characterize the T cell response in AAD. It could be concluded the T cells in AAD patientsrespond to three immunodominant epitopes on the 21-hydroxylase.
In addition, this thesis aims to gain updated data on comorbidities, replacement therapy,autoantibody profiles, and metabolic factors in AAD. A cohort of 660 AAD patients was studiedand it was found that AAD patients are prone to develop other autoimmune conditions. AAD isone of three main disease components Autoimmune Polyendocrine Syndrome type 1 (APS-1),a rare disorder caused by mutations in the AutoImmune REgulator gene (AIRE) that can bepotentially fatal without timely diagnosis. Screening for autoantibodies against interferon-ω,interferon-α4, and interleukin-22 revealed four new APS-1 patients among the AAD cohort,confirmed by the presence of disease causing mutations in the AIRE gene.
Cancer Associated Retinopathy (CAR) is a paraneoplastic phenomenon arising as aconsequence to an autoimmune response triggered by a malignant neoplasm present in the body.This disease is devastating and it is valuable to identify new biomarkers associated with CAR,not least from a tumor diagnostic perspective. In this thesis, a patient with osteosarcoma andCAR was studied and by screening of a proteom array, and the novel CAR autoantigen Arylhydrocarbon receptor interacting protein-like 1 (AIPL1) was identified.
In conclusion, this thesis covers studies on T cell and B cell responses in AAD. Moreover,it includes an update on clinical and immunological characterisation of AAD patients. Finally,a novel autoantigen in CAR was identified and proposed as a diagnostic marker for theparaneoplastic syndrome.
Frida Dalin, Department of Medical Sciences, Akademiska sjukhuset, Uppsala University,SE-75185 Uppsala, Sweden.
© Frida Dalin 2015
ISSN 1651-6206ISBN 978-91-554-9392-9urn:nbn:se:uu:diva-265276 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-265276)
List of Papers
This thesis is based on the following papers, which are referred to in the text by their Roman numerals.
I Dawoodji A, Chen JL, Shepherd D, Dalin F, Tarlton A, Alimohammadi
M, Penna-Martinez M, Meyer G, Mitchell AL, Gan EH, Bratland E, Bensing S, Husebye ES, Pearce SH, Badenhoop K, Kämpe O, Cerundo-lo V. High frequency of cytolytic 21-hydroxylase-specific CD8+ T cells in autoimmune Addison's disease patients. J Immunol. 2014 Sep 1;193(5):2118-26.
II Frida Dalin, Grazyna Adamus, Sufang Yang, Eva Landgren, Josefine Palle, Åsa Hallgren, Britt-Marie Frost MD, Therése Hugosson, Nils Landegren, Daniel Eriksson, Sten Andréasson, Khalid F Tabbara, Olle Kämpe, Mohammad Alimohammadi. Aryl Hydrocarbon Receptor-Interacting Protein-Like 1 (AIPL1) - a Novel Retinaspecific Autoantigen in Cancer-Associated Retinopathy. In revision Ophthalmology
III Frida Dalin, Gabriel Nordling Eriksson, Per Dahlqvist, Åsa Hallgren, Jeanette Wahlberg, Olov Ekwall, Per Olcén, Stefan Söderberg, Johan Rönnelid, Ola Winquist, Sergiu Catrina, The Swedish Addison Registry Study Group*, Anna-Lena Hulting, Olle Kämpe, and Sophie Bensing.
Clinical and immunological characteristics of Autoimmune Addison’s disease in Sweden: a nationwide multicenter analysis of 660 patients. Manuscript
IV Frida Dalin, Daniel Eriksson, Gabriel Nordling Eriksson, Nils Landegren, Åsa Hallgren, Per Dahlqvist, Jeanette Wahlberg, Olov Ek-wall, Per Olcén, Stefan Söderberg, Ola Winquist, Sergiu Catrina, Johan Rönnelid, The Swedish Addison Registry Study Group*, Anna-Lena Hulting, Mohammad Alimohammadi, Gerli Rosengren Pielberg, Sophie Bensing, and Olle Kämpe. Identification and characterization of APS-1 patients by autoantibody screening of the Swedish Addison Registry. Manuscript.
Reprints were made with permission from the respective publishers.
Contents
Introduction ..................................................................................................... 9 The Immune System and Immunological Tolerance .................................. 9 Autoimmunity and Autoimmune Disease ................................................ 10 Autoimmune Polyendocrine Syndrome type 1 ......................................... 10 Autoimmune Addison’s disease ............................................................... 11 Paraneoplastic Syndromes and Cancer-Associated Retinopathy ............. 13
Current investigatons ..................................................................................... 15 Aims .......................................................................................................... 15 Materials and Methods ............................................................................. 15
In vitro Transcription and Translation (Manuscripts I, II, III) ............ 15 Protein Array Screening (Manuscript I) .............................................. 16 Immunofluorescence Staining (Manuscript I) ..................................... 16 Recall Assay (Paper I) ......................................................................... 16 Ex vivo ELISPOT ................................................................................ 17 Granzyme B Assay (Paper I) ............................................................... 18 Statistical Analysis (Manuscript II) ..................................................... 18
Results and Discussion ............................................................................. 18 Cytolytic 21-hydroxylase-specific CD8+ T cells mediate adrenal cortex destruction in autoimmune Addison's disease patients (Paper I) ................................................................................. 18 Identification of Aryl Hydrocarbon Receptor-Interacting Protein-Like 1 (AIPL1) as a novel retina-specific autoantigen in Cancer-Associated Retinopathy (Manuscript I) .................................. 22 Clinical, immunological and epidemiological characteristics of Autoimmune Addison’s disease (Manuscript II) ............................ 24 Identification of subclinical APS-1 patients within the Swedish Addison registry (Manuscript III) ........................................................ 29
Conclusions ................................................................................................... 32
Future Perspectives ....................................................................................... 33
Summary of Thesis in Swedish ..................................................................... 34
Acknowledgement ......................................................................................... 37
References ..................................................................................................... 39
Abbreviations
17α-OH 17α-hydroxylase 21-OH 21-hydroxylase AAD Autoimmune Addison’s disease AADC Aromatic L-amino acid decarboxylase ACTH Adrenocorticotropic hormone AIRE Autoimmune regulator APC Allophycocyanin (fluorescent dye) APS-1 Autoimmune polyendocrine syndrome type 1 AR Autoimmune retinopathy BCR B cell receptor BSA Bovine serum albumin CAR Cancer-associated retinopathy CRH Corticotropin releasing hormone CY3 Indocarbocyanine (fluorescent dye) ELISPOT Enzyme-linked immunosorbent spot ERG Electroretinography FITC Fluorescein isothiocyanate (fluorescent dye) GAD65 Glutamate decarboxylase 65 HPA Hypothalamic-pituitary-adrenal IA-2 Islet cell antigen 2 ICS Intracellular staining IFN Interferon IL Interleukin ITT In vitro transcription and translation KCNRG Potassium channel regulator MHC Major histocompatibility complex mTECs Medullary thymic epithelial cells NALP5 NACHT, leucine rich repeat and PYD containing 5 PBMC Peripheral blood mononuclear cell PerCP Peridinin chlorophyll (fluorescent dye) RLBA Radio-ligand binding assay SOX10 SRY (sex determining region Y)-box 10 T1DM Type 1 diabetes mellitus T2DM Type 2 diabetes mellitus TCR T cell receptor TMB (3,3′,5,5′-tetramethylbenzidine)
TPH Tryptophan hydroxylase TPO Thyroid peroxidase TGM2 Tissue transglutaminase 2
9
Introduction
Autoimmune diseases include a spectrum of more than 80 different disor-ders, estimated to affect 5% of the population. Many diseases that were pre-viously considered to be idiopathic are in fact suspected to have autoimmune origin [1]. Autoimmune diseases usually have complex inheritance patterns and occur as a consequence when several factors such as heredity and envi-ronment interact [2-4]. These parameters make it difficult to study autoim-mune diseases, which hamper the development of new therapies [5]. Auto-immune disease models are therefore very important for understanding the pathogenesis of this broad disease group.
The Immune System and Immunological Tolerance The human immune system is a complex organization of several types of cells and molecules that protect us from invading microorganisms [6]. It is divided into two fundamentally different types of responses: the innate re-sponse and the adaptive response. The innate response is the non-specific, first line defence against an invading microorganism and occurs to the same extent every time the pathogen is encountered, whereas the adaptive re-sponse is improved upon repeated exposure to the same infection [6]. The adaptive immune system consists of several cell types, although the two major ones are bone marrow-derived B cells and thymus-derived T cells, both originating from a common pluripotent lymphoid progenitor. These cells possess highly specialized receptors for a given antigen: the soluble or transmembrane B cell receptor (BCR) and the transmembrane T cell receptor (TCR) [7].
Once a functional TCR is produced, immature T cells undergo a number of selection events. First, the TCR is tested for its antigen-recognition ability against molecules present in the thymus. Medullary thymic epithelial cells (mTECs) present self-antigens on their surface, bound to major histocompat-ibility complex (MHC) molecules. This promiscuous expression of self-antigens is regulated by the autoimmune regulator gene (AIRE)[8]. Only lymphocytes whose receptors bind weakly to self-antigens bound to the body’s MHC molecules are selected to survive and develop further. This process is termed positive selection. Lymphocytes with receptors that bind strongly to self-antigens are selected to undergo apoptosis; this is termed
10
negative selection and is an important in preventing self-reactive lympho-cytes from damaging the body’s own tissue. In this way, immunological tolerance to self-antigens is established [6, 9, 10].
Autoimmunity and Autoimmune Disease Despite the mechanism of self-tolerance, some self-reactive lymphocytes do mature and cause autoimmunity when the organism begins producing an ab-normal immune response to its own tissue. This generally involves self-reactive cytolytic T cells and autoantibody-producing self-reactive B cells [11]. Some-times pathogens express antigens that resemble host molecules, either by chance or as an immune escape mechanism. This can result in activated lymphocytes cross-reacting with self-antigens and consequently induction of autoimmunity. If autoimmunity leads to pathological changes such as tissue destruction, the con-dition is recognized as an autoimmune disease [5].
Autoimmune diseases are often classified as either organ-specific, where the autoimmune reaction is localized to an isolated organ, or systemic, where the autoimmunity is widespread in the body [12]. The inflammation or cellu-lar destruction occurring in an autoimmune disease may be caused directly by T cells or indirectly by T cells that sustain autoantibody responses.
The presence of serum autoantibodies have proven to be valuable markers for many autoimmune diseases and is an important tool in diagnosing and predicting diseases development since autoantibodies may be present years before clinical symptoms [13]. Such biomarkers can be measured when the autoantigen is known. It is therefore of great interest to identify autoantigens associated with various autoimmune diseases.
Autoimmune Polyendocrine Syndrome type 1 Autoimmune Polyendocrine Syndrome type 1 (APS-1) is a multi-organ auto-immune disorder caused by mutations in the AutoImmune REgulator gene (AIRE), located on chromosome 21q22 [14]. The AIRE protein is a unique transcriptional regulator, located in the nucleus of mTECs, where it pro-motes the promiscuous expression of thousands of peripheral organ-specific autoantigens in the thymus during T cell maturation. This is a step critical for the induction of immunological self-tolerance [9, 15, 16].
The hallmarks of APS-1 are hypoparathyroidism, chronic mucocutaneous candidiasis, and adrenal failure. However, there are today more than 20 dif-ferent autoimmune endocrine and non-endocrine manifestations that have been identified in APS-1 [17, 18]. Patients with APS-1 develop autoantibod-ies to multiple tissue-specific antigens [17, 19, 20] and the presence of these autoantibodies is used to recognize and diagnose the syndrome (table 1). The
11
majority of patients have three to five different disease manifestations [21, 22].
A hallmark of APS-1 is chronic mucocutaneous candidiasis. Circulating neutralizing anti-cytokine autoantibodies, such as several type I interferon isoforms, e.g. interferon-ω (IFNω) and interferon-α4 (IFNα4) [23, 24] and IL-17/IL-22 [25-27] are present in the vast majority of APS-1 patients. It is hypothesized that autoantibodies against the interleukins may be responsible for the chronic mucocutaneous candidiasis infections present in almost all APS-1 patients, since these cytokines play key roles in inducing defensins, small proteins important for the immune response to pathogens on mucosal surfaces [27]. Measuring these serum autoantibodies is an important diag-nostic tool, as they are found in almost all APS 1 patients before manifesta-tions of most symptoms [20, 28]. Table 1. Disease component associated with different target autoantigens in APS-1.
Disease component Related autoantigen Adrenal failure 21-OH, 17α-OH, SCC Gonadal failure SCC Hypoparathyroidism NALP5 Diabetes GAD65, IA-2 Respiratory symptoms KCNRG Intestinal symptoms TPH Hepatitis AADC Alopecia TH Vitiligo SOX9, SOX10 ?1) IFNω, IFNα4 Chronic Mucocutaneous Candidiasis? IL-22, IL-17A
1) No known associated disease component.
Autoimmune Addison’s disease Autoimmune Addison’s disease (AAD) is a primary adrenal failure caused by autoimmune destruction of the adrenal cortex. The disease was first de-scribed by Thomas Addison in 1855 [29]. The principal symptoms of Addi-son’s disease are fatigue, reduced appetite, weight loss, salt cravings, and hyperpigmentation of the skin due to elevated adrenocorticotropic hormone levels [30, 31]. An important hallmark of AAD is the presence of autoanti-bodies targeting the steroidogenic 21-hydroxylase (21-OH), an intracellular key enzyme in cortisol and aldosterone production, selectively expressed in the adrenal cortex [32]. Circulating adrenal cortex autoantibodies can be detected several years before clinical onset; they therefore serve as predic-tive markers for the disease [33]. At diagnosis, >90% of Addison patients have 21-OH autoantibodies, and the prevalence slowly declines with disease
12
duration, due to decreased size of the remaining adrenal cortex tissue [34, 35]. Less than 0.5% of the general population have 21-OH autoantibodies. However, 50% of people with autoantibodies to 21-OH never develop Addi-son’s disease [33], the autoantibodies may thus not be the causative factor of AAD but rather an indication of a T cell mediated destruction of the adrenal cortex.
As a consequence of the adrenal gland destruction, patients become inca-pable of producing sufficient steroid hormones such as glucocorticoids (e.g. cortisol) and mineralocorticoids (e.g. aldosterone) [36]. The adrenal glands constitute together with hypothalamus and pituitary glands the hypothalam-ic–pituitary–adrenal (HPA) axis, mainly regulated by the corticotrophin-releasing hormone (CRH) and the adrenocorticotropic hormone (ACTH) (figure 1). The HPA-axis activity is controlled by the negative feedback of cortisol on the hypothalamus and the pituitary gland (figure 1). It helps con-trol reactions to stress and regulates many body processes such as mood, sexuality, and metabolism [37]. In Addison’s patients, the lack of endoge-nous cortisol production impairs the negative feedback loop and results in elevated CRH and ACTH levels.
Figure 1. The normal function of the Hypothalamic-Pituitary-Adrenal (HPA) axis and its regulation. The hypothalamus releases CRH which upregulates the secretion of ACTH by the anterior pituitary. ACTH in turn acts on the adrenal cortex to stimu-late cortisol production. Cortisol has a negative feedback effect on both the hypo-thalamus and the anterior pituitary to decrease CRH and ACTH release respectively.
Untreated, AAD can lead to the life-threatening condition Addisonian crisis, characterized by very low blood pressure and ensuing coma [30]. Treatment of Addison patients should include glucocorticoid and mineralocorticoid replacement, whilst the benefit of androgen substitution remains debatable [36, 38]. The ideal glucocorticoid substitution therapy would mimic the en-dogenous circadian cortisol rhythm, although this is not possible to obtain with the current oral replacement regimens that generate periods of supra- and infra-physiological cortisol levels [39]. Overtreatment with glucocorti-coids has been connected to negative metabolic consequences e.g. impaired
13
glucose tolerance, dyslipidaemia, hypertension, and osteoporosis [40-42]. It is believed that today’s non-optimized replacement therapy may contribute to the increased mortality and morbidity reported in AAD [43].
Autoimmune Addison’s disease is frequently associated with other non-adrenal organ-specific autoimmune disorders, most commonly hypothyroid-ism, and the patient then suffers from an autoimmune polyendocrine syn-drome, APS [31, 38]; non-endocrine conditions such as vitiligo and celiac disease, however, are also typically associated with Addison’s disease [44]. APS-1 is the monogenic disorder mentioned above whereas APS-2 is a much more common and genetically complex entity characterized by two or more concurrent autoimmune endocrinopathies, e.g. AAD, autoimmune hypo- or hyperthyroidism and type 1 diabetes [45-47]. The associated diseases are often connected with the presence of autoantibodies against a specific anti-gen (table 1). After 21-OH autoantibodies, anti-thyroid peroxidase (TPO), associated with hypothyrodism is most frequently found in AAD patients. The third most common autoantibodies are parietal cell antibodies, associat-ed with pernicious anemia. However, autoantibodies against many other autoantigens have been associated with Addison’s disease and APS-1, e.g. 17α-OH, SCC, and GAD65 [47]. Other autoantibodies are considered to be specific for APS-1, such as IFNω, IFNα4, and IL-22.
The Swedish Addison Registry is the world’s largest collection of clinical data and blood from patients with AAD. To estimate the total number of Addison patients in Sweden, individuals with both a diagnosis of AAD and on combination treatment with hydrocortisone/cortisone acetate and fludro-cortisone were identified in 2013 through the Swedish National Patient Reg-istry and the Swedish Prescribed Drug Registry [48]. With this approach, the number of patients with AAD in Sweden was estimated to be 1305. Assum-ing this number is still valid, the Swedish Addison Registry covers approxi-mately 57% of the AAD patients in Sweden.
Paraneoplastic Syndromes and Cancer-Associated Retinopathy Paraneoplastic syndromes are rare, tumour-associated disorders that are not a direct effect of the tumour itself. These conditions are either mediated by humoral factors, such as hormones or cytokines secreted by the tumour, or mediated by an immunological response against the tumour [49, 50]. Immu-nologically mediated paraneoplastic syndromes arise when the immune re-sponse against the tumour cross-reacts with healthy tissue where the same or cognate antigens are normally expressed. The location where the autoantigen is expressed will therefore determine the type of organ or tissue involved [51]. Some autoimmune diseases arise almost exclusively as paraneoplastic
14
manifestations whereas many autoimmune diseases can occur either in pres-ence or in absence of an underlying malignancy [52]. Paraneoplastic retinop-athies can be divided into cancer-associated retinopathy (CAR) and mela-noma-associated retinopathy (MAR). These are autoimmune retinal degen-erative disorders, occurring in a small subset of patients with systemic can-cer. CAR is characterized by sudden, bilateral, and painless vision loss and photosensitivity, and is associated with anti-retinal autoantibodies [53-55].
Paraneoplastic syndromes occur in 7-10% of patients with malignant neo-plasms. The number of patients with visual symptoms of CAR and undiag-nosed malignancy are continuously increasing with growing physician awareness and improved diagnostic tools and criteria. Nevertheless, it still remains a significant diagnostic and therapeutic challenge [56]. The latency time from the presentation of vision loss to detection of tumour may be sev-eral years; hence the presence of autoantibodies may serve as a predictive marker of an underlying neoplasm [53, 57].
CAR is most commonly associated with breast cancer, followed by lung and gynaecological cancers. Different malignancies are often associated with specific CAR antigens, e.g. breast cancer is associated with autoantibodies against α-enolase and transductin-α, while cervical cancers are associated with anti-recoverin antibodies. There are also a number of unknown antigens with known molecular weights identified by Western blotting, that are asso-ciated with different malignancies. Several different methods are used to evaluate the presence of anti-retinal antibodies: immunofluorescence testing, western blot testing, and ELISA testing. However, testing for antiretinal antibodies are not currently standardized among laboratories, hence the re-sults are not always concordant between different testing sites [58]. There is a need for standardized testing methods and identification of new CAR asso-ciated biomarkers.
15
Current investigations
Aims
I. Investigate the mechanisms behind the T cell mediated destruction of the adrenal glands in autoimmune Addison's disease. II. Identify new retinal structure/s and autoantigen/s in patients suffering from cancer-associated retinopathy.
III. Study patients with autoimmune Addison’s disease to gain deeper in-sights into clinical and immunological features to provide upgraded data regarding autoimmune comorbidities, replacement therapy, autoantibody profiles, and metabolic factors.
IV. Investigate the frequency o undiagnosed cases of APS-1 in the Swedish Addison Registry, and evaluate the prediction value of currently available diagnostic serological and genetic markers for the diagnosis of APS-1.
Materials and Methods In vitro Transcription and Translation (Manuscripts I, II, III) Full-length cDNA clones were used to produce recombinant 35S-radiolabeled protein using the TnT® system (Promega). Trichloroacetic acid precipitation and scintillation counting measured the 35S-methionine incorporation. 20 000 counts per minute (cpm) of the 35S-radiolabeled recombinant protein were incubated with patient or blood donor serum over night at 225 rpm at 4oC. The immune reactions were transferred to 96-well microtiter filtration plates (Millipore) and were incubated with protein-A for 45 min at 225 rpm at 4oC. On each plate, a positive standard, an APS-1 patient with previously known high titre autoantibodies against the antigen of interest, and a negative stand-ard, 4% bovine serum albumin (BSA), were included. As positive standard in manuscript I, a commercial anti-AIPL1 antibody was used. The radioac-tivity of the immunoprecipitated material was measured with a liquid scintil-lation counter (Wallac 1450 MicroBeta, Perkin-Elmer). The autoantibody detection results were expressed as an index: [(cpm in the unknown sample –
16
cpm in negative standard) ÷ (cpm in the positive standard – cpm in the nega-tive standard) * 100]. The upper limit of the normal range was defined as the mean index value for blood donors plus 3 standard deviations (SD) for man-uscripts II and III, and 4 SD in manuscript I. For 21-OH autoantibody analy-sis in manuscript II and III, the limit for positive index values was based on maximum accuracy cut-off estimations from a recent interlaboratory study [59].
Protein Array Screening (Manuscript I) A human proteome microarray (HuProtTM) (CDI Laboratories, Baltimore, MD), containing 19 394 proteins covering approximately 80% of the protein coding genes, was screened with serum from a patient with CAR and osteo-sarcoma. The proteome array was incubated with serum in dilution 1:2000 for 90 min, at 50 rpm at room temperature followed by washing and addition of secondary detection anti-human Alexa 647 conjugated antibody (Life Technologies) in the dark for 50 min at 50 rpm at room temperature. The array was then washed and scanned using a CapitalBio LuxScan HT24 mi-croarray scanner.
Immunofluorescence Staining (Manuscript I) Human retinal cryosections were fixed with 4% paraformaldehyde for 10 min followed by permeabilization, and blocking at room temperature. The sections were incubated with human serum overnight at 4°C, followed by incubation with secondary anti-human antibody conjugated to Alexa 488. After washing and blocking, a rabbit anti-AIPL1 antibody in dilution 1:100 was added and the sections were incubated for 1 h, followed by washing and incubation with a secondary anti-rabbit antibody conjugated to Alexa 594. The sections were washed again and mounting reagent with DAPI was add-ed. The immunofluorescent labelling was evaluated using an Olympus Fluoview1000 confocal microscope.
Porcine eye cryosections were fixed in ice-cold acetone for 2 min fol-lowed by blocking at room temperature. The sections were incubated with human serum in dilution 1:400 overnight at 4°C. Antibody binding was de-tected using a FITC-conjugated goat anti-human antibody in dilution 1:100. The sections were washed, and mounting reagent with DAPI was added. The immunofluorescent labelling was evaluated under a Zeiss LSM 510 META confocal microscope.
Recall Assay (Paper I) Overlapping 18-aa peptides spanning the entire 21-OH protein dissolved in DMSO (40 µg/ml) were used to stimulate patient PBMCs either individually
17
or as a combined pool. Six million cells per well were stimulated with 1 µg/ml 21-OH peptide pool in RH-10 medium with IL-7. On day 3, RH-10 medium containing IL-2 was added and the cells were cultured with this medium until day 13, when they were washed to remove IL-2 and incubated in RH-10 overnight. On day 14, 200 000 cells were pulsed with individual peptides or the pool of peptides for 5 h in presence of brefeldin A, followed by washing and staining for the surface markers CD8 conjugated to APC-H7, CD4 conjugated to PerCP, CD3 conjugated to V500, CD107 conjugated to PE together with LIVE/DEAD viability marker for 30 min on ice. The cells were then washed and fixed in fixation solution overnight at 4°C. The following day, the cells were permeabilized and stained for intracellular for cytokine secretion with FITC conjugated antibody against IFNγ for 30 min on ice. Cells were washed and re-suspended in FACS buffer for acquisition by flow cytometry (FACSCantoII) and data were analysed using FlowJo (TreeStar).
Figure 2. Overview of the recall assay.
Ex vivo ELISPOT (Paper I) Plates were coated at 4°C overnight with capture antibody in coating buffer and the next day washed followed by 1 h blocking at 37°C. The plates were washed again and 500 000 cells per well were incubated with 10 µg/ml indi-vidual peptide or peptide pool were incubated overnight at 37°C. The plates were washed before adding biotin detection antibody, then incubated for 2 h at 37°C followed by washing and adding of Streptavidin-ALP and the plates were then incubated for 1 h at room temperature. Finally, the plates were washed before developing by adding substrate solution for 10 min in dark at room temperature. Rinsing the plates thoroughly with cold water stopped development and the plates were left to dry completely before spots were counted using an AID EliSpot Reader.
18
Granzyme B Assay (Paper I) 20 000 CD8+ T cells were co-cultured with 40 000 target cells from the HLA-A2–expressing tumour line NCI H295 overnight followed by meas-urement of Granzyme B secretion in the supernatant using ELISA. The ELI-SA plates were coated overnight with anti-Granzyme B antibody before ad-dition of the supernatant and then incubated overnight at 4°C. The next day, the plates were washed and incubated with biotin antibody, followed by ad-dition of ExtrAvidin Peroxidase. The plates were developed using TMB solution.
Statistical Analysis (Manuscript II) T-test was used when investigating correlations between a non-categorical variable with a categorical variable. Chi2-tests and Fisher’s exact test were used when investigating correlation between proportions. Linear logistic regression was used for analysis of 21-OH autoantibody index and disease duration. For comparisons of the SAR and MONICA cohorts linear logistic regression was used for BMI and binary logistic regression for hypertension, hyperlipidaemia, and type 2 diabetes. Glucocorticoid replacement dose (mg/day) was recalculated to hydrocortisone equivalent dose (mg/m2/day) accordingly:
All the above statistical analyses were performed using the SAS v9.4, Prism v6 and SPSS 23.0 software. Significance was accepted if p<0.05 for all anal-yses in this study.
Results and Discussion Cytolytic 21-hydroxylase-specific CD8+ T cells mediate adrenal cortex destruction in autoimmune Addison's disease patients (Paper I) The target antigen for autoimmune Addison’s disease (AAD), 21-OH, is exclusively expressed in the adrenal cortex [60]. The mechanisms behind
19
destruction of the adrenal glands in AAD remain unclear and it is speculated that T cells mediate the disease.
In this study, we wanted to identify the immunodominant T cell epitope/s on the 21-OH protein. To do this, peripheral blood monocytes (PBMCs) were isolated from 20 patients with AAD with high 21-OH antibody titres and 7 healthy controls. The PBMCs were incubated with overlapping 18-aa peptides spanning the entire 21-OH protein for 13 days in a recall assay. At day 15, the IFN$ secretion from CD4+ and CD8+ T cells respectively was quantified via FACS. Expanding the patient’s PBMCs with a pool of pep-tides revealed that the majority of the 20 Addison’s disease patients had high frequency of 21-OH–specific CD8+ and CD4+ T cells, compared with healthy controls (figure 3, upper panels). Expanding PBMCs with each indi-vidual peptide revealed that the majority of patients had CD8+ T cells capa-ble of recognizing 21-OH337–354 and/or 21-OH428–445, and CD4+ T cells rec-ognizing the peptide 21-OH 207–224 (figure 3, lower panels). Frequencies of 21-OH specific T cell responses were also compared between patients diag-nosed with Addison’s disease for up to 20 years previously, revealing that although the magnitude of the 21-OH–specific T cell response decreased over time, it is still present even at the point when the adrenal cortex is hypothesized to be completely de-stroyed. This indicates a continuous stimulation of 21-OH–specific T cells, perhaps mediated by ectopic expression of 21-OH in other tissues or by pathogens expressing a cross-reactive epitope.
Figure 3. Upper panels: The production of IFN$ from (A) CD8+ and (B) CD4+ T cells in response to 21-OH in AAD patients or healthy controls under re-stimulation
20
with the 21-OH peptide pool or unstimulated (DMSO). Lower panels: 21-OH pep-tide pool bulk cultured PBMCs were re-stimulated with individual peptides, DMSO or the 21-OH pool and stained intracellularly for IFNγ secretion.
By ex vivo ELISPOT assay using patients' PBMCs we could show that the same immunodominant peptides as in the recall assay evoked T cell respons-es. The ELISPOT assay provides a better understanding of the magnitude of 21-OH–specific T cells as the recall assay indicates a memory T cell re-sponse rather than representing the actual frequency of 21-OH–specific T cells. By ex vivo ELISPOT assay, we identified which epitope on the immu-nodominant peptide 21-OH337–354 evoked the strongest T cell response by incubating PBMCs with overlapping 9-mers spanning the 18 amino acid long peptide. The 9-mer 21-OH342–350 (LLNATIAEV) elicited the strongest response and was subsequently used for sorting CD8+ T cell clones to per-form functional studies. The T cell clones were stained with HLA-A2 and HLA-B8 tetramers loaded with the peptides 21-OH342–350 and 21-OH428–435 to confirm the specificity and HLA restriction of the T cell response.
The CD8+ T cells were tested in VITAL and Granzyme B assay and for their ability to recognize 21-OH–expressing cells. In the VITAL assay, the CD8+ T cells recognized target cells pulsed with 21-OH342–350 peptide, but not with irrelevant peptides. These data provide direct evidence that the T cell populations derived from PBMCs of Addison’s disease patients contain HLA-A2–restricted CD8+ T cells that recognize the peptide 21-OH342–350. In the Granzyme B assay, the 21-OH342–350–specific HLA-A2–restricted CD8+ T cell clone was co-cultured with the HLA-A2–expressing tumour cell line NCI H295, and an increase of Granzyme B secretion in the supernatant, as measured by ELISA was observed. Addition of HLA-A2–blocking antibody BB7.2 reduced NCI H295 recognition (figure 4).
Figure 4. Granzyme B secretion detectable in the supernatant by ELISA when 21-OH specific T cells are cocultured with NCI H295 cells. The response is increased when exogenous 21-OH(342-350) peptide is pulsed to target cells before coculture and reduced when the A2 blocking antibody, BB7.2 is added
21
The ability of 21-OH–specific CD8+ T cell clones to recognize endogenous 21-OH protein may be directly linked to the progression of Addison’s dis-ease. Since only a minority of individuals with 21-OH autoantibodies devel-op Addison’s disease within 5 years [33], future investigations of T cell re-sponses in these individuals and the ability to predict disease progression are of major importance, and may possibly result in diagnostic and therapeutic opportunities. A long-term vision would also be to use the understanding of T cell response initiation in Addison’s disease to induce responses against adrenal cortex cancer; a 21-OH–positive tumour with very poor prognosis that often afflicts younger people.
22
Aryl Hydrocarbon Receptor-Interacting Protein-Like 1 (AIPL1) is a novel retina-specific autoantigen in Cancer-Associated Retinopathy (Manuscript I) Paraneoplastic syndromes constitute symptoms of tissue damage remote from a malignant neoplasm present in the body and in many cases the para-neoplastic symptoms arise due to an immunological reaction. In this study, we identify a novel autoantigen in a patient with cancer-associated retinopa-thy (CAR) concomitant with malignant osteosarcoma. The patient was a previously healthy 15-year-old female patient with two years’ history of pain in her right calcaneus and simultaneous development of increased photosen-sitivity and bilateral rapid visual loss. Optical coherence tomography (OCT) showed a macular atrophy with photoreceptor thinning and electroretino-gram (ERG) demonstrated a reduction of rod-cone function by more than 90% as compared to a normal response some months before the patient be-came blind. When the patient’s symptoms in her calcaneus were investigated with magnetic resonance imaging (MRI) and subsequent biopsy, it revealed a small-cell tumour that was later characterized as osteosarcoma.
By screening a proteome array with patient serum, the retina and pineal gland specific protein Aryl hydrocarbon receptor interacting protein-like 1 (AIPL1) was identified as the immune target. Evaluation of AIPL1 as an autoantigen in CAR using a radioligand-binding assay with 73 healthy blood donors and 100 other patient sera from patients with CAR and different ma-lignancies, or autoimmune retinopathy (AR), revealed that AIPL1 is an un-common autoantigen among CAR/AR patients. To localize the target antigen in the retina, human retinal tissue was immunostained with serum from the patient with osteosarcoma and CAR, a healthy control, and a commercial monoclonal AIPL1 antibody. The patient’s autoantibodies were directed towards the interphotoreceptor matrix in the retina. Confocal images showed that the AIPL1 is targeted in rod photoreceptor cells (figure 5). These results are consistent with previous studies showing that AIPL1 is expressed in both developing cone and rod photoreceptors but restricted to rod photoreceptors in the adult retina, suggesting that AIPL1 is essential for development of normal rod and cone photoreceptor function but is only required for rod sur-vival in the adult [61]. The symptoms of CAR depend on whether the rod or cone function is primarily disrupted. In this case, with the major autoantigen being AIPL1, expressed in rods, the signs are often constriction of the visual field with impaired dark adaptation [62].
23
Figure 5. Indirect immunofluorescence staining of human retinal tissue with serum sample from the CAR patient (dilution 1:50) and AIPL1 antibody (dilution 1:100). Green represents secondary antibody conjugated to Alexa 488, staining structures recognized by the patient serum. Red represents secondary antibody conjugated to Alexa 594 staining structures recognized by the monoclonal anti-AIPL1. Blue repre-sents DAPI stained nuclei. CAR serum stains the rod photoreceptors and also cones (green). The anti-AIPL1 antibody stains mainly rods (red). The merged image shows an overlap between anti-AILP1 antibody and CAR patient serum (yellow). The synaptic region of photoreceptor cells is strongly labelled. Scale bars represent 20 µm in all sub-panels.
Homozygous or compound heterozygous mutations in the gene encoding AIPL1 on chromosome 17p13 cause the disease Leber congenital amaurosis 4 (LCA4) (OMIM# 604393). Patients affected by LCA4 are diagnosed at birth or in the first months of life with nystagmus, severely impaired vision or blindness, and absence of rod and cone function, as demonstrated by ERG [63]. Sohocki et al. identified the disease causing nonsense mutation W278X in AIPL1 in 2000 and it is estimated that mutations in the AIPL1 gene might account for approximately 20% of LCA in isolated populations [64] and 7% worldwide [65]. The human aryl hydrocarbon receptor interacting protein-like 1 is named for its similarity to the human aryl hydrocarbon receptor-interacting protein (AIP) and belongs to the FK506-binding protein (FKBP) family. In the retina, it acts as a specific chaperone for phosophodiesterase-6 (PDE6), a visual effector enzyme, which is important in phototransduction. Mutations in AIPL1 result in destabilization of PDE6 and thereby disturbed phototransduction, resulting in LCA4 [66].
24
Intriguingly, AIPL1 is only expressed in photoreceptor cells and the pine-al gland. This co-expression suggests that the pineal gland is a potential tar-get for autoimmune disorders. The pineal gland has a diurnal release of mel-atonin, resetting circadian rhythm. However, the exact function of AIPL1 in this tissue remains elusive although it is not unlikely that AIPL1 mutations may also give rise to pineal-associated abnormalities in LCA4 patients [64].
When the same protein is deficient due to either genetic mutations or au-toimmune destruction, the autoimmune disorder is considered to be a pheno-copy of the genetic disorder, meaning that CAR could be an acquired auto-immune phenocopy of the inherited disease LCA4. In phenocopic disorders, the shared molecular targets are often important proteins with key functions in various tissues. In this context, this suggests that AIPL1 is a plausible key protein in the retina.
In retrospect, it is apparent that the patient presented with CAR long be-fore the tumour was diagnosed. This is nothing unique, paraneoplastic symp-toms precede the diagnosis of the neoplasm more often than is usually rec-ognised [53, 67]. This empathises the challenge to recognize when a symp-tom are associated with an underlying malignancy and the importance of the identification of paraneoplastic autoantibodies also from a tumour diagnostic perspective. Facilitating earlier diagnosis and treatment could in this particu-lar case perhaps at least partly have saved the patient’s vision. The retinal damage is to the largest extent considered to be irreversible; nevertheless, tumour removal may reduce the disease’s progression, underlining the im-portance of early diagnosis.
Clinical and immunological characteristics of Autoimmune Addison’s disease (Manuscript II) In order to make an updated characterization of Addison patients, a compre-hensive clinical and immunological study was conducted on 660 patients with autoimmune Addison’s disease in the multi-centre Swedish Addison Registry (SAR). This is the largest patient cohort available internationally. Because autoimmune Addison’s disease (AAD) is a rare disease, deeper insights into clinical and immunological features are needed to optimize monitoring of these patients. Clinical data from registration forms were evaluated and presence of autoantibodies against 13 different autoantigens were determined in all patients.
All patients were screened for autoantibodies against associated antigens 21-OH, 17α-OH, SCC, SOX10, AADC, TPO, GAD65, TGM2, deamidated gliadin, parietal cells, and the APS-1 indicative autoantigens IFNω, IFNα4, and IL-22. Patients with any of the APS-1 suggestive autoantibodies were also screened for conventional APS-1 associated autoantibodies; TH, TPH, KCNRG, CPY1A2, and NALP5.
25
This study validates some previous findings and provides upgraded data regarding gender differences, autoimmune comorbidities, replacement thera-py, autoantibody profiles and, for the first time in such a large cohort, data on metabolic factors.
The proportion of 21-OH autoantibody positive patients was 83% and these patients had significantly shorter disease duration (mean 15.6 years) than the 21-OH negative patients (mean 23.9 years). Our study also revealed a significant negative correlation (r=-0.26) between 21-OH autoantibody levels and disease duration (figure 6). A similar correlation has previously been seen in a smaller cohort (n=94) of Norwegian AAD patients [34]. Nev-ertheless, 21-OH autoantibodies were often detectable in the circulation a long time after AAD diagnosis, which is intriguing since it is believed that the autoimmune attack results in complete adrenal gland destruction and with this, the presence of the autoantigen.
Figure 6. Prevalence of 21-OH autoantibodies among 660 patients in the SAR co-hort. Panel A shows 21-OH autoantibody index value plotted against disease dura-tion (R2=0.068; p<0.0001). Panel B shows the proportion of 21-OH autoantibody positive patients in different disease duration intervals.
We could confirm that AAD more commonly affects women than men [34] and the mean age at diagnosis was significantly higher for women, 36.7 years, than for men with 31.1 years (95% CI: 3.45-7.76; p<0.0001), which is concurrent with previous studies [34, 68, 69]. This might mirror the fact that the disease actually develops later in life among women, or that the time from that the disease develops until diagnosis is longer in women, which in turn might be a consequence of women being more neglected in the healthcare system than men, as similar tendencies have been observed for patients with acute coronary syndrome [70, 71]. Another hypothesis might be that women are protected from developing autoimmunity by pregnancy [72, 73]. The majority of the patients in the SAR (62%) had one or more associated autoimmune disease with a woman to man ratio of 1.03:0.64 (p<0.0001). The finding that women with AAD have a higher degree of au-toimmune comorbidity as compared to men with AAD is in line with the trend seen in the general population [74, 75]. Hypothyrodism (p<0.0001),
26
hyperthyrodism (p=0.0028), hypogonadism (p=0.0015), and alopecia (p=0.0454) had significantly higher frequencies among women than men in our study. The female preponderance of hypogonadism may partly have two explanations; women get more obvious symptoms than men, i.e. amenor-rhea, in contrast to the more diffuse symptom erectile dysfunction [46] or that the gonadal autoantigen is more protected in men because of the blood-testis barrier.
The most common concomitant diseases were hypothyroidism, followed by type 1 diabetes, pernicious anaemia, and hyperthyroidism. In the SAR cohort, 264/660 (40%) had hypothyroidism, which is a considerably higher proportion than in previous studies, with frequencies between 6% and 22% [35, 46]. In the general population, hypothyroidism affects less than 1% of people [35, 75]. TPO autoantibodies were seen in 57% of the SAR popula-tion with a significant association to hypothyroidism. These results suggest that TPO antibodies and development of hypothyroidism should be moni-tored regularly in Addison’s patients. The frequency of type 1 diabetes (T1DM) was 11% among the Addison patients, concurrent with previous studies (7-12%) [35, 46], which could be compared to a prevalence of T1DM in the general population of 0.5% [76]. However, we did not see any significantly higher frequency of T1DM among men, which was to be ex-pected as it is more common among men in the general population [77]. GAD65 autoantibodies were detected in 23% of the cohort, showing a clear association to T1DM (p<0.0001). The prevalence of pernicious anaemia was 11% among the AAD patients in the SAR, similar to previous results [46]. However, 33% of the AAD patients were positive for parietal cell autoanti-bodies, strongly associated with pernicious anaemia. This suggests that the actual number of patients suffering from pernicious anaemia is under report-ed in our patient cohort and that B12 deficiency should be monitored regu-larly in Addison’s patients. Surprisingly, only 17 (2.6%) had transglutami-nase (TGM2) autoantibodies in this cohort, and 15 (2.3%) had diagnosed celiac disease. These numbers are not significantly higher than the estimated 1% of the general Western world population having celiac disease [78-80]. However, only five of the patients with celiac disease diagnosis had TGM2 autoantibodies suggesting that 12 patients with TGM2 autoantibodies do not yet have a diagnosis; hence the actual number of celiac disease patients is rather 27 (4.1%). This number is still lower than 5.6-7.9% in previous stud-ies [35, 81]. A higher prevalence of celiac disease was previously shown among AAD patients with APS-1 [82], which may partly explain the low incidence of celiac disease in this study, where APS-1 patients are excluded.
There was an overall discrepancy between serum autoantibody findings and clinical diagnoses of corresponding disease. For example, 374 patients had TPO autoantibodies but only 264 had hypothyroidism and 153 patients had GAD65 autoantibodies while only 71 had T1DM. GAD65 autoantibod-ies in non-diabetic patients might indicate subclinical insulitis, which in time
27
can progress to overt T1DM [83, 84], suggesting these patients should be regularly followed up.
We also analysed if the autoimmune components were diagnosed before, at the same time as, or after AAD diagnosis. This revealed that the median time from AAD diagnosis to hypothyroidism diagnosis was 0 years (Q1-Q3; -1;5 years) (p <0.0001), possibly reflecting that the patient is screened for numerous diseases when getting the AAD diagnosis. Another explanation would be that an increase in TSH due to low cortisol is misinterpreted as hypothyroidism. From AAD diagnosis to pernicious anaemia diagnosis, the median time was +10 years (Q1-Q3; 0;22 years) (p<0.0001). No other asso-ciated diseases showed any significant differences in time of diagnosis. However, hypertension, hyperlipidaemia, and type 2 diabetes take 15-20 years to develop; mirroring these may be unfavourable side effects from long-term glucocorticoid substitution therapy.
Glucocorticoid replacement was predominantly covered by regular hy-drocortisone (89%) with a mean dose of 28.0 mg/day (std dev 8.5), divided into one (12.9%), two (49.5%), three (33.0%), or four (3.3%) daily doses. The mean daily hydrocortisone dose was slightly higher than 15-25 mg/day, which is generally the recommendation today [85]. When glucocorticoid doses were normalized to body surface and recalculated to hydrocortisone equivalent dose, the mean dose was 14.8 mg/m2/day (std dev 4.4). Very little data exist on the actual doses used in AAD cohorts, however, in a Norwe-gian study, the hydrocortisone equivalent daily dose, recalculated from corti-sone acetate, was 32.4 mg/day [35] and a south African study reports a me-dian hydrocortisone dose of 23.8 mg/day [41].
The hydrocortisone equivalent dose was compared for the different glu-cocorticoid replacement regimens, where patients receiving hydrocortisone modified release tablet Plenadren® (n=52) had a 2.8 mg/m2/day (95% CI: 1.56-4.04; p<0.0001) lower mean total hydrocortisone equivalent dose, and patients with prednisolone (n=3) had a 5.1 mg/m2/day (95% CI 0.07-10.08; p=0.047) lower hydrocortisone equivalent dose than patients on regular hy-drocortisone substitution (figure 7).
28
Figure 7. Hydrocortisone equivalent doses in AAD patients with different glucocor-ticoid replacement regimens.
The ideal glucocorticoid replacement therapy aims to mimic the endogenous cortisol rhythm [86, 87], and minimize unfavourable metabolic side effects and decrease in quality of life. This means low cortisol levels in the evening, gradually rising during the night and peaking in the morning before waking, followed by a decrease during the day [88]. However, today’s conventional glucocorticoid replacement therapy inadequately imitates this pattern. A too high hydrocortisone dose and non-physiological peaks in cortisol levels are associated with unfavourable metabolic side effects and decrease in quality of life [86, 87, 89], and previous studies show that hydrocortisone doses exceeding 30 mg/day more severely impairs the quality of life compared to lower doses [86, 90, 91].
Mineralocorticoid substitution was utilized by 89% of the patients with a mean dose of 0.093 mg/day (std dev: 0.054 mg/day). The recommended daily dose is 0.05-0.2 mg [31] but data suggests that 0.2 mg of fludrocorti-sone is needed to maintain adequate sodium and water balance [92]. This indicates that the patients on mineralocorticoid substitution in the SAR are receiving too low doses for appropriate effects. The patients not receiving fludrocortisone substitution (11%) had a 1.2 mg/m2/day higher mean hydro-cortisone equivalent dose as compared to those receiving fludrocortisone replacement (95% CI: 0.16-2.30; p=0.0240). The long-term effects of being without or on too low a mineralocorticoid dose and compensate this with higher glucocorticoid dose are not known. No correlation between daily fludrocortisone dose and incidence of hypertension, hyperlipidemia, or type 2 diabetes could be seen.
When analysing the occurrence of metabolic factors, 553 patients in the AAD cohort (25-75 years of age) were compared to the population-based control cohort MONICA (MONItoring of Trends and Determinants of CAr-diovascular Disease) (n=3627) from northern Sweden. The SAR cohort con-
Hydro
corti
sone
Plenad
ren®
Cortiso
ne ace
tate
Predniso
lone0
10
20
30
40H
ydro
corti
sone
equ
ival
ent d
ose
mg/
m2 /d
ayp<0.0001
p=0.047
29
tained a significantly higher proportion of women (59.7%) than the control cohort (50.9%) (p<0.0001). The comparison revealed that despite higher than recommended average doses of glucocorticoid replacement therapy among the Addison’s patients, they did not have a higher risk of type 2 dia-betes or hyperlipidaemia, usually more frequent in AAD patients with gluco-corticoid replacement than in the general population. Furthermore, the AAD patients had a lower risk of hypertension than the control population, and BMI was significantly lower in patients with AAD compared with the con-trol subjects, mean difference 1.27 kg/m2* (95% CI 0.86-1.68 kg/m2) (*adjusted for age and sex), p<0.0001).
Long-term glucocorticoid treatment is generally associated with increased BMI [93, 94]. The finding that the Addison patients in this study have lower BMI may concur with the lower risk of having hypertension among the Ad-dison cohort. The results are contradictory to some previous studies, show-ing that patients with primary adrenal insufficiency have an increased mor-tality risk with cardiovascular diseases as the leading cause of death [68, 95]. However, in these studies the cause of death is not considered and the cardi-ovascular mortality among AAD patients may be partly explained by that the cause of death may be acute cortisol deficiency during the intensive care rather than from cardiovascular causes. In summary; all these results on met-abolic effects indicate that the patients in the SAR are receiving an adequate glucocorticoid replacement therapy.
Identification of subclinical APS-1 patients within the Swedish Addison registry (Manuscript III) Autoimmune Addison’s disease is a major disease component in autoim-mune polyendocrine syndrome type 1 (APS-1). Therefore, following the comprehensive autoantibody testing in the previous study, patients in the Swedish Addison registry positive for any of the autoantibodies against IFNω, IFNα4, or IL-22 were classified as possible undiagnosed APS-1 pa-tients and were further screened for conventional APS-1 associated autoanti-bodies against TH, TPH, KCNRG, CYP1A2, and NALP5. In this study, also the APS-1 patients in the SAR were included, resulting in 677 Addison pa-tients. Of these patients, 34 had IFNω, IFNα4, and/or IL-22 autoantibodies. Fourteen of these (41%) had autoantibodies against more than one of these three autoantigens (figure 8). Thirteen of the 34 AAD patients with APS-1 autonatibodies fulfilled the clinical criteria for APS-1 and already had an established APS-1 diagnosis. Four of the remaining 25 suspected new APS-1 patients had disease-causing mutations in AIRE. This leaves 21 patients with IFNω, IFNα4, and/or IL-22 autoantibodies but without APS-1.
30
Figure 8. Autoantibody profiles for 677 patients in the Swedish Addison Registry. Overview of the different patient subgroups from the antibody screening and se-quencing is shown to the left. Diagram over the autoantibody profiles of the suspect-ed (n=25) and known (n=13) APS-1 patients is shown to the right. Each circle repre-sents one of the three autoantibodies IFNω, IFNα4, and IL-22. The number outside the circle represents patients without any of the autoantibodies. Known APS-1 pa-tients are indicated by *.
It can be speculated that the patients with IFN/IL-22 autoantibodies without disease causing AIRE mutations or clinical manifestations indicative for APS-1 may be more closely related to each other than the rest of the Addi-son cohort or other APS-1 patients. However, by clustering all genetic vari-ants of AIRE, this does not seem to hold. Since previous studies show that these autoantibodies are present in almost all APS-1 patients before manifes-tations of most symptoms [20, 28], it is important to keep these patients un-der observation for signs of late onset APS-1.
Patients with autoantibodies against IFNω, IFNα4, or IL-22 and no clini-cal symptoms typical of APS-1 and no identified disease causing AIRE mu-tations may indicate an intermediate form of APS-1, and these patients should be kept under observation. Since APS-1 patients can present with a wide range of clinical phenotypes, we suggest that all Addison patients should be screened for IFN/IL-22 antibodies. IFNω antibodies are almost always present before APS-1 is diagnosed and persistent, making them ex-cellent disease markers, particularly for early or atypical patients who have not developed the full APS I picture [24]. Furthermore, APS-1 diagnosis should especially be considered in patients developing AD in childhood, and when a new patient is diagnosed, all siblings should be investigated for APS-1 manifestations and AIRE mutations [96].
Recent findings by Oftedal et al suggests that mono-allelic mutations in the PHD1 domain of AIRE can partially or completely disrupt the AIRE protein in a dominant negative manner giving milder phenotypes with later onset as compared to classical autosomal-recessive inherited APS-1. This can be explained by that some wild-type AIRE tetramers are still formed inducing some degree of self-tolerance [97]. However, the sequencing of
31
these patients in our study did not reveal such mono-allelic variants in the PHD1 domain (exon 8 and 9 of AIRE).
32
Conclusions
I. Both CD4+ and CD8+ 21-OH specific T cells are abundant in a large pro-portion of Addison’s patients both in vivo and ex vivo assays. CD8+ T cells are able to lyse 21-OH+ target cells, consistent with a potential mechanism for disease pathogenesis.
II. AIPL1 is a novel paraneoplastic retinal autoantigen, a finding demon-strating autoimmune-mediated CAR to be a phenocopy of the inherited dis-order Leber congenital amaurosis type 4 (LCA4). We suggest that assess-ment of anti-retinal antibodies against AIPL1 might be useful not only to diagnose the retinal disease, but also to guide the clinician in searching for an underlying undiagnosed tumour or metastatic spread of cancer.
III. Patients with AAD are prone to develop autoimmune comorbidities; therefore these patients should be carefully monitored. The replacement therapy needs to be adjusted to follow recommended doses, however, the AAD patients do not have more unfavourable metabolic side effects than the control population.
IV. Four hitherto novel APS-1 patients in the SAR were identified by auto-antibody screening. The patients had typical APS-1 autoantibody profiles, AIRE mutations, and fulfilled the clinical criteria for APS-1. The importance of screening for APS-1 autoantibodies should not be underestimated when it comes to diagnosing possible APS-1 that is not clinically overt. Since the clinical presentation of APS-1 is variable and often takes years to develop, patients having autoantibodies indicative for APS-1 should be further exam-ined for mutations in the AIRE gene to help earlier diagnosis and allowing treatments to prevent serious complications.
33
Future Perspectives
This thesis describes the use of autoimmune model diseases to gain a better understanding of how and why autoimmunity arises. In the first paper, we identified three immunodominant epitopes in patients with autoimmune ad-renal insufficiency. A better understanding of this initiation may open up for development of a T cell mediated immunotherapy against the aggressive cancer adrenal cortical carcinoma, a rare endocrine malignancy with limited treatment options and poor prognosis [98, 99]. Today’s treatment is based on complete surgical removal of the adrenals, but in half of the patients this is impossible due to metastasized disease [100]. Research has been conducted on immunotherapies in adrenal cortical carcinoma; however, adoptive T cell therapy is a new approach that not yet has been approved for this type of cancer [101, 102]. Using an adrenal cortex carcinoma cell-line (NCI-H295), the cytolytic ability of the T cells selected for the immunodominat epitope of 21-OH could be investigated ex vivo. This experiment can be considered as an initial study to assess the possibility of using adoptive T cell therapy in adrenal cortex carcinoma.
Deeper studies on paraneoplastic syndromes in general and the AIPL1 an-tigen in particular is needed for increased understanding of the T cell re-sponse to AIPL1 protein and other known paraneoplastic antigens, e.g. to explore if any of the autoantigens associated with CAR have shared epitopes. It is also of great interest to evaluate more CAR patient serum samples on proteome arrays to identify possible new autoantigens. It would also be of interest to study the role of AIPL1 in the pineal gland. Moreover, development of a multiplex testing with known paraneoplastic autontigens on a custom-made protein array would improve diagnostics for parane-oplastic syndromes.
Longitudinal studies on a group of patients with Addison’s disease would be needed to monitor how several factors such as 21-OH antibody titres, T cell responses, and development of associated endocrinopathies, alter with disease progression.
34
Summary of Thesis in Swedish
Vårt immunförsvar består av ett komplext nätverk som skyddar oss mot an-grepp från bakterier, virus och parasiter. Två centrala celltyper som ingår i immunförsvaret är antikroppsproducerande B-celler, och cytotoxiska T-celler.
En av immunförsvarets grundläggande uppgifter är att urskilja vilka strukturer som tillhör den egna kroppen och inte angripa dessa. Trots allt sker dessa självdestruktiva angrepp emellanåt och fenomenet kallas autoim-munitet och en sådan aktivering av immunförsvaret kan ge upphov till auto-immuna sjukdomar. De brukar delas in i organspecifika, där specifika organ angrips av immunförsvaret, eller systemiska, där autoantigener är spridda i hela kroppen. Trots att autoimmuna sjukdomar är vanligt förekommande är mycket lite känt om de mekanismer som har betydelse för dess uppkomst.
Ett kännetecken för många autoimmuna sjukdomar är att B-celler produ-cerar antikroppar mot särskilda antigen. Genom att upptäcka dessa antikrop-par kan man diagnostisera många sjukdomar och det är därför viktigt att identifiera dessa sjukdomsmarkörer som är specifikt associerade med olika autoimmuna sjukdomar. För att underlätta sökandet av nya sjukdomsmarkö-rer använder man sig av olika sjukdomsmodeller.
Den här avhandlingen beskriver studier av sjukdomsmekanismer i tre olika autoimmuna sjukdomar; Autoimmun kronisk binjurebarksvikt (Ad-disons sjukdom), Autoimmunt Polyendokrint Syndrom typ 1 (APS-1) och det paraneoplastiska syndromet Cancerassocierad Retinit (CAR).
Addisons sjukdom är kronisk autoimmun binjurebarksvikt där de flesta patienter har autoantikroppar mot enzymet 21-hydroxylas (21-OH) som ut-trycks i binjurebarken. Binjurebarken producerar i normala fall livsnödvän-diga stresshormoner, bl.a. kortisol, vilket gör att personer med Addisons sjukdom behöver ersätta den uteblivna kroppsegna produktionen av kortisol - ett livsviktigt hormon för flera vitala processer i kroppen. Utebliven korti-solsubstitution innebär det livshotande tillståndet Addisonkris med extremt låg blodtryck och koma.
Jag har studerat T-cellers roll i Addisons sjukdom, då det är mest troligt att det är T-celler som initierar sjukdomen och i sin tur aktiverar antikropps-producerande B-celler. Genom att ingående förstå mekanismerna bakom hur Addisons sjukdom initieras skulle man i förlängningen kunna använda auto-reaktiva 21-OH specifika T-celler som vaccinering vid binjurebarkscancer där de då skulle attackera tumörcellerna. Jag kunde se att T-cellerna känner
35
igen tre olika immundominanta epitoper på 21-OH proteinet och att aktive-rade T-celler från patienter med Addisons sjukdom känner igen och attacke-rar binjurebarkcancerceller som uttrycker 21-OH.
Vidare har jag studerat en stor kohort Addisonpatienter som är registre-rade i Svenska Addison Registret (SAR) för att karaktärisera denna grupp av patienter och undersöka en mängd olika faktorer så som associerade auto-immuna sjukdomar, förekomst av autoantikroppar, metabola biverkningar av medicinering mm. Jag valde ut 660 patienter ur SAR som hade autoimmun binjurebarksvikt, vilket gör den här studien till världens störta studie av Ad-disonpatienter, och kunde bl.a. se att 83 % hade antikroppar mot 21-OH som sakta avtog med ökad sjukdomstid. De vanligaste associerade sjukdomarna är hypotyroidism följt av typ 1-diabetes och perniciös anemi. Vid jämförelse med en kontrollkohort verkade Addisonpatienterna inte ha högre risk för att utveckla typiska metabola biverkningar såsom högt BMI, hypertension, hy-perlipidemi och typ 2-diabetes av sin medicinering, vilket var oväntat ef-tersom dessa är förväntade effekter vid långvarig glukokortikoidsubstitue-ring.
En extrem form av autoimmunitet är den ovanliga sjukdomen APS-1 där patienten utvecklar många olika organspecifika autoimmuna sjukdomskom-ponenter. APS-1 orsakas av att en viktig gen, Autoimmun Regulator (AIRE), som kontrollerar immunförsvarets förmåga att inte angripa kroppsegna strukturer är muterad och resulterar i ett icke fungerande protein. Patienter med APS-1 har höga nivåer autoantikroppar mot ett antal kända vävnadsspe-cifika autoantigen. Samma autoantikroppar som identifierats hos APS-1 patienter kan också förekomma vid isolerade former av de autoimmuna sjukdomskomponenterna. Det har tidigare visat sig att olika signaleringsmo-lekyler som interferon-ω, interferon-α4 och interleukin-22 är bra biomarkö-rer för sjukdomen då majoriteten av alla APS-1-patienter har antikroppar mot dessa proteiner. Därför kan de användas för att hitta misstänkta odia-gnostiserade APS-1-patienter bland riskgrupper; såsom patienter med auto-immun kronisk binjurebarksvikt (Addisons sjukdom). Genom att undersöka 677 patienter ur Svenska Addisonregistret för autoantikroppar mot dessa karaktäristiska APS-1-markörer, studera deras kliniska symtom och sekven-sera AIRE genen för att hitta sjukdomsframkallande mutationer identifierade jag fyra nya APS-1-patienter i kohorten. Det är av stor vikt att APS-1-patienter får rätt behandling, då obehandlad sjukdomen är livshotande.
Vidare har jag studerat en patient med det paraneoplastiska syndromet cancerassocierad retinit. Paraneoplastiska syndrom uppstår när det finns en tumör i kroppen som uttrycker en mängd olika antigen vilket resulterar i en massiv aktivering av immunförsvaret i syfte att bekämpa tumören. Av en slump kan tumören uttrycka proteiner som normalt även uttrycks någonstans i övriga kroppen, och immunförsvaret kan då missriktas och attackera kroppsegen vävnad med höga nivåer autoantikroppar som följd. Dessa auto-antikroppar finns ofta i cirkulationen innan tumören upptäcks och kan såle-
36
des vara användbara biomarkörer ur ett tumördiagnostiskt perspektiv. Patien-ten som jag har studerat var en ung flicka med hastigt försämrad syn och smärtor i en fot. Först när patienten var helt blind upptäcktes att orsaken till smärtan var skelettcancer i hälbenet, vilket resulterade i diagnosen osteosar-kom och cancerassocierad retinit. Jag identifierade det retinaspecifika anti-genet AIPL1 som målet för patientens autoimmuna angrepp på retina. Man kan spekulera i att med en mer utvecklad diagnostik av CAR som inkluderar den nya biomarkören AIPL1, hade en tidigare diagnos getts och patientens syn delvis kunnat räddas.
Sammanfattningsvis bidrar denna avhandling till ökad förståelse för bakomliggande mekanismer vid autoimmuna sjukdomar och kan ligga till grund för vidare studier som syftar till att utveckla diagnostik och terapier.
37
Acknowledgements
The work of this thesis was mainly carried out at the Department of Medical Sciences at Uppsala University. However, many experiments were per-formed at Center for Molecular Medicine at the Karolinska Institute and at the Weatherall Institute of Molecular Medicine at University of Oxford, where I have met incredibly talented and helpful people. I would like to ex-press my gratitude to all of them, with a special thanks to the following peo-ple: The heads of the Department of Medical Sciences at Uppsala University Lars Rönnblom and Eva Tiensuu Jansson. My main supervisor Mohammad Alimohammadi for always having a solu-tion to every problem and for great discussions. Thank you for encourage me to do things in good time and for still being patient with me when I did not. My co-supervisor professor Olle Kämpe for his great knowledge, generosi-ty, cooking expertise, and for taking such good care of the research group. My other co-supervisor Sophie Bensing for her strong support, realistic advice and for all the inspiring discussions and good suggestions. Åsa Hallgren, for knowing everything and for all the help and support over the years. Thank you for always having time and patience, and for being a great friend also outside of the lab. My fellow commuters (and PhD colleagues) at Karolinska: Nils Landegren and Daniel Eriksson. Thank you for making every day in the lab easier and for all the great scientific and not-so-scientific discussions during fikas. I would also like to thank the new lab colleagues at CMM for being so friend-ly and for welcoming us to the new lab. Friends and co-workers in Uppsala: Kerstin Ahlgren, Anna Lobell, Anna-Stina Sahlqvist, Brita Ardesjö-Lundgren, Magnus Isaksson, Fredrik Rorsman, Håkan Hedstrand, and Gerli Rosengren Pielberg. I would also like to give a special thanks to all other members of CKMF 2 and 3 for al-ways being friendly and helpful.
38
I would like to thank all my collaborators, with a special thanks to Vincenzo Cerundolo and Amina Dawoodji and the other co-workers at the Weath-erall Institute of Molecular Medicine in Oxford for welcoming me to the lab and making the stay there so worthwhile. Per Dahlquist and Marie Eriks-son in Umeå for great statistical inputs. Per Morten Knappskog and Eystein Husebye in Bergen, and pediatric oncologists Josefine Palle and Britt-Marie Frost in Uppsala. All of my co-authors and of course the patients. My friends outside the lab Matilda, for being the best BFF ever, and for our new routine, dubious thai buffet lunches. Ebba and Sofie for great BSF dis-cussions. Hanna, Annica, Mie, and Freddis for making life easier. My parents Kerstin and Rolf for their love, encouragement, and invaluable life advice. My sisters and brothers Linda, Elin, Linus, and Andreas for always being there when needed. A special thank to Linda (again), not only for being my slightly older and considerably wiser sister but also for being my personal statistician with extremely flexible accessibility. I would also like to thank the Winquist family for their incredible hospitality. My amazing son Frans, thank you for filling my life with happiness. To my joy, my friend, my companion, my love Thomas, thank you for everything! And last but not least I would like to thank Darwin, for being the best dog in the world.
39
References
1. Najjar, S., et al., Immunology and epilepsy. Rev Neurol Dis, 2008. 5(3): p. 109-16.
2. Shepshelovich, D. and Y. Shoenfeld, Prediction and prevention of autoimmune diseases: additional aspects of the mosaic of autoimmunity. Lupus, 2006. 15(3): p. 183-90.
3. Ballotti, S., F. Chiarelli, and M. de Martino, Autoimmunity: basic mechanisms and implications in endocrine diseases. Part I. Horm Res, 2006. 66(3): p. 132-41.
4. Morahan, G. and L. Morel, Genetics of autoimmune diseases in humans and in animal models. Curr Opin Immunol, 2002. 14(6): p. 803-11.
5. Su, M.A. and M.S. Anderson, Aire: an update. Curr Opin Immunol, 2004. 16(6): p. 746-52.
6. Delves, P.J. and I.M. Roitt, The immune system. First of two parts. N Engl J Med, 2000. 343(1): p. 37-49.
7. Goodnow, C.C., et al., Cellular and genetic mechanisms of self tolerance and autoimmunity. Nature, 2005. 435(7042): p. 590-7.
8. Zumer, K., K. Saksela, and B.M. Peterlin, The mechanism of tissue-restricted antigen gene expression by AIRE. J Immunol, 2013. 190(6): p. 2479-82.
9. Anderson, M.S., et al., Projection of an immunological self shadow within the thymus by the aire protein. Science, 2002. 298(5597): p. 1395-401.
10. Klein, L., et al., Positive and negative selection of the T cell repertoire: what thymocytes see (and don't see). Nat Rev Immunol, 2014. 14(6): p. 377-91.
11. Castro, C. and M. Gourley, Diagnostic testing and interpretation of tests for autoimmunity. J Allergy Clin Immunol, 2010. 125(2 Suppl 2): p. S238-47.
12. Fridkis-Hareli, M., Immunogenetic mechanisms for the coexistence of organ-specific and systemic autoimmune diseases. J Autoimmune Dis, 2008. 5: p. 1.
13. Betterle, C., et al., Autoimmune adrenal insufficiency and autoimmune polyendocrine syndromes: autoantibodies, autoantigens, and their applicability in diagnosis and disease prediction. Endocr Rev, 2002. 23(3): p. 327-64.
14. Finnish-German, A.C., An autoimmune disease, APECED, caused by mutations in a novel gene featuring two PHD-type zinc-finger domains. Nat Genet, 1997. 17(4): p. 399-403.
15. Hollander, G.A. and P. Peterson, Learning to be tolerant: how T cells keep out of trouble. J Intern Med, 2009. 265(5): p. 541-61.
16. Abramson, J. and Y. Goldfarb, AIRE: From promiscuous molecular partnerships to promiscuous gene expression. Eur J Immunol, 2015.
17. Alimohammadi, M., et al., Autoimmune polyendocrine syndrome type 1 and NALP5, a parathyroid autoantigen. N Engl J Med, 2008. 358(10): p. 1018-28.
18. Perheentupa, J., Autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy. J Clin Endocrinol Metab, 2006. 91(8): p. 2843-50.
19. Ekwall, O., et al., Identification of tryptophan hydroxylase as an intestinal autoantigen. Lancet, 1998. 352(9124): p. 279-83.
40
20. Husebye, E.S., et al., Clinical manifestations and management of patients with autoimmune polyendocrine syndrome type I. J Intern Med, 2009. 265(5): p. 514-29.
21. Ahonen, P., et al., Clinical variation of autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy (APECED) in a series of 68 patients. N Engl J Med, 1990. 322(26): p. 1829-36.
22. Myhre, A.G., et al., Autoimmune polyendocrine syndrome type 1 (APS I) in Norway. Clin Endocrinol (Oxf), 2001. 54(2): p. 211-7.
23. Oftedal, B.E., et al., Radioimmunoassay for autoantibodies against interferon omega; its use in the diagnosis of autoimmune polyendocrine syndrome type I. Clin Immunol, 2008. 129(1): p. 163-9.
24. Meager, A., et al., Anti-interferon autoantibodies in autoimmune polyendocrinopathy syndrome type 1. PLoS Med, 2006. 3(7): p. e289.
25. Puel, A., et al., Autoantibodies against IL-17A, IL-17F, and IL-22 in patients with chronic mucocutaneous candidiasis and autoimmune polyendocrine syndrome type I. J Exp Med, 2010. 207(2): p. 291-7.
26. Kisand, K., et al., Chronic mucocutaneous candidiasis in APECED or thymoma patients correlates with autoimmunity to Th17-associated cytokines. J Exp Med, 2010. 207(2): p. 299-308.
27. Ahlgren, K.M., et al., Increased IL-17A secretion in response to Candida albicans in autoimmune polyendocrine syndrome type 1 and its animal model. Eur J Immunol, 2011. 41(1): p. 235-45.
28. Wolff, A.S., et al., Anti-cytokine autoantibodies preceding onset of autoimmune polyendocrine syndrome type I features in early childhood. J Clin Immunol, 2013. 33(8): p. 1341-8.
29. Addison, T., On The Constitutional And Local Effects Of Disease Of The Supra-Renal Capsules. London, 1855.
30. Arlt, W., Adrenal insufficiency. Clin Med, 2008. 8(2): p. 211-5. 31. Arlt, W. and B. Allolio, Adrenal insufficiency. Lancet, 2003. 361(9372): p.
1881-93. 32. Winqvist, O., et al., Two different cytochrome P450 enzymes are the adrenal
antigens in autoimmune polyendocrine syndrome type I and Addison's disease. J Clin Invest, 1993. 92(5): p. 2377-85.
33. Coco, G., et al., Estimated risk for developing autoimmune Addison's disease in patients with adrenal cortex autoantibodies. J Clin Endocrinol Metab, 2006. 91(5): p. 1637-45.
34. Myhre, A.G., et al., Autoimmune adrenocortical failure in Norway autoantibodies and human leukocyte antigen class II associations related to clinical features. J Clin Endocrinol Metab, 2002. 87(2): p. 618-23.
35. Erichsen, M.M., et al., Clinical, immunological, and genetic features of autoimmune primary adrenal insufficiency: observations from a Norwegian registry. J Clin Endocrinol Metab, 2009. 94(12): p. 4882-90.
36. Bancos, I., et al., Diagnosis and management of adrenal insufficiency. Lancet Diabetes Endocrinol, 2015. 3(3): p. 216-26.
37. De Kloet, E.R., et al., Brain corticosteroid receptor balance in health and disease. Endocr Rev, 1998. 19(3): p. 269-301.
38. Husebye, E.S., et al., Consensus statement on the diagnosis, treatment and follow-up of patients with primary adrenal insufficiency. J Intern Med, 2014. 275(2): p. 104-15.
41
39. Bjornsdottir, S., et al., Circadian hormone profiles and insulin sensitivity in patients with Addison's disease: a comparison of continuous subcutaneous hydrocortisone infusion with conventional glucocorticoid replacement therapy. Clin Endocrinol (Oxf), 2014.
40. Giordano, R., et al., Metabolic and cardiovascular profile in patients with Addison's disease under conventional glucocorticoid replacement. J Endocrinol Invest, 2009. 32(11): p. 917-23.
41. Ross, I.L., et al., Increased cardiovascular risk in South African patients with Addison's disease. Horm Metab Res, 2013. 45(12): p. 905-10.
42. Lovas, K., et al., Glucocorticoid replacement therapy and pharmacogenetics in Addison's disease: effects on bone. Eur J Endocrinol, 2009. 160(6): p. 993-1002.
43. Oksnes, M., et al., Continuous subcutaneous hydrocortisone infusion versus oral hydrocortisone replacement for treatment of addison's disease: a randomized clinical trial. J Clin Endocrinol Metab, 2014. 99(5): p. 1665-74.
44. Michels, A.W. and G.S. Eisenbarth, Autoimmune polyendocrine syndrome type 1 (APS-1) as a model for understanding autoimmune polyendocrine syndrome type 2 (APS-2). J Intern Med, 2009. 265(5): p. 530-40.
45. Husebye, E. and K. Lovas, Pathogenesis of primary adrenal insufficiency. Best Pract Res Clin Endocrinol Metab, 2009. 23(2): p. 147-57.
46. Fichna, M., et al., Screening for associated autoimmune disorders in Polish patients with Addison's disease. Endocrine, 2010. 37(2): p. 349-60.
47. Soderbergh, A., et al., Prevalence and clinical associations of 10 defined autoantibodies in autoimmune polyendocrine syndrome type I. J Clin Endocrinol Metab, 2004. 89(2): p. 557-62.
48. Bjornsdottir, S., et al., Drug prescription patterns in patients with Addison's disease: a Swedish population-based cohort study. J Clin Endocrinol Metab, 2013. 98(5): p. 2009-18.
49. Pelosof, L.C. and D.E. Gerber, Paraneoplastic syndromes: an approach to diagnosis and treatment. Mayo Clin Proc, 2010. 85(9): p. 838-54.
50. Darnell, R.B. and J.B. Posner, Paraneoplastic syndromes affecting the nervous system. Semin Oncol, 2006. 33(3): p. 270-98.
51. Byrne, K.T. and M.J. Turk, New perspectives on the role of vitiligo in immune responses to melanoma. Oncotarget, 2011. 2(9): p. 684-94.
52. Lleo, A., et al., Definition of human autoimmunity--autoantibodies versus autoimmune disease. Autoimmun Rev, 2010. 9(5): p. A259-66.
53. Adamus, G., Autoantibody targets and their cancer relationship in the pathogenicity of paraneoplastic retinopathy. Autoimmun Rev, 2009. 8(5): p. 410-4.
54. Volpe, N.J. and J.F. Rizzo, 3rd, Retinal disease in neuro-ophthalmology: paraneoplastic retinopathy and the big blind spot syndrome. Semin Ophthalmol, 1995. 10(3): p. 234-41.
55. Klaas, J.P., et al., Adult-onset opsoclonus-myoclonus syndrome. Arch Neurol, 2012. 69(12): p. 1598-607.
56. Braithwaite, T., A. Vugler, and A. Tufail, Autoimmune retinopathy. Ophthalmologica, 2012. 228(3): p. 131-42.
57. Adamus, G., Latest updates on antiretinal autoantibodies associated with vision loss and breast cancer. Invest Ophthalmol Vis Sci, 2015. 56(3): p. 1680-8.
58. Gordon, L.K., Paraneoplastic Syndromes in Neuro-Ophthalmology. J Neuroophthalmol, 2015. 35(3): p. 306-14.
42
59. Falorni, A., et al., Determination of 21-hydroxylase autoantibodies: inter-laboratory concordance in the Euradrenal International Serum Exchange Program. Clin Chem Lab Med, 2015.
60. Winqvist, O., F.A. Karlsson, and O. Kampe, 21-Hydroxylase, a major autoantigen in idiopathic Addison's disease. Lancet, 1992. 339(8809): p. 1559-62.
61. van der Spuy, J., et al., Predominant rod photoreceptor degeneration in Leber congenital amaurosis. Mol Vis, 2005. 11: p. 542-53.
62. Keltner, J.L., et al., Management and monitoring of cancer-associated retinopathy. Arch Ophthalmol, 1992. 110(1): p. 48-53.
63. Cremers, F.P., J.A. van den Hurk, and A.I. den Hollander, Molecular genetics of Leber congenital amaurosis. Hum Mol Genet, 2002. 11(10): p. 1169-76.
64. Sohocki, M.M., et al., Mutations in a new photoreceptor-pineal gene on 17p cause Leber congenital amaurosis. Nat Genet, 2000. 24(1): p. 79-83.
65. Sohocki, M.M., et al., Prevalence of AIPL1 mutations in inherited retinal degenerative disease. Mol Genet Metab, 2000. 70(2): p. 142-50.
66. Majumder, A., et al., Interaction of aryl hydrocarbon receptor-interacting protein-like 1 with the farnesyl moiety. J Biol Chem, 2013. 288(29): p. 21320-8.
67. Heckenlively, J.R. and H.A. Ferreyra, Autoimmune retinopathy: a review and summary. Semin Immunopathol, 2008. 30(2): p. 127-34.
68. Bergthorsdottir, R., et al., Premature mortality in patients with Addison's disease: a population-based study. J Clin Endocrinol Metab, 2006. 91(12): p. 4849-53.
69. Nerup, J., Addison's disease--clinical studies. A report fo 108 cases. Acta Endocrinol (Copenh), 1974. 76(1): p. 127-41.
70. Herlitz, J., et al., Characteristics and outcome among women and men transported by ambulance due to symptoms arousing suspicion of acute coronary syndrome. Med Sci Monit, 2002. 8(4): p. CR251-6.
71. Maynard, C., et al., Influence of sex on the use of cardiac procedures in patients presenting to the emergency department. A prospective multicenter study. Circulation, 1996. 94(9 Suppl): p. II93-8.
72. Hazes, J.M., et al., Pregnancy and the risk of developing rheumatoid arthritis. Arthritis Rheum, 1990. 33(12): p. 1770-5.
73. Kung, A.W. and B.M. Jones, A change from stimulatory to blocking antibody activity in Graves' disease during pregnancy. J Clin Endocrinol Metab, 1998. 83(2): p. 514-8.
74. Cooper, G.S. and B.C. Stroehla, The epidemiology of autoimmune diseases. Autoimmun Rev, 2003. 2(3): p. 119-25.
75. Jacobson, D.L., et al., Epidemiology and estimated population burden of selected autoimmune diseases in the United States. Clin Immunol Immunopathol, 1997. 84(3): p. 223-43.
76. Jansson, S.P., D.K. Andersson, and K. Svardsudd, Prevalence and incidence rate of diabetes mellitus in a Swedish community during 30 years of follow-up. Diabetologia, 2007. 50(4): p. 703-10.
77. Gale, E.A. and K.M. Gillespie, Diabetes and gender. Diabetologia, 2001. 44(1): p. 3-15.
78. Ludvigsson, J.F., et al., Support for patients with celiac disease: A literature review. United European Gastroenterol J, 2015. 3(2): p. 146-59.
79. Green, P.H., B. Lebwohl, and R. Greywoode, Celiac disease. J Allergy Clin Immunol, 2015. 135(5): p. 1099-106; quiz 1107.
43
80. Kelly, C.P., et al., Advances in diagnosis and management of celiac disease. Gastroenterology, 2015. 148(6): p. 1175-86.
81. Myhre, A.G., et al., High frequency of coeliac disease among patients with autoimmune adrenocortical failure. Scand J Gastroenterol, 2003. 38(5): p. 511-5.
82. Betterle, C., et al., Celiac disease in North Italian patients with autoimmune Addison's disease. Eur J Endocrinol, 2006. 154(2): p. 275-9.
83. Tuomi, T., et al., Antibodies to glutamic acid decarboxylase and insulin-dependent diabetes in patients with autoimmune polyendocrine syndrome type I. J Clin Endocrinol Metab, 1996. 81(4): p. 1488-94.
84. Gylling, M., et al., ss-cell autoantibodies, human leukocyte antigen II alleles, and type 1 diabetes in autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy. J Clin Endocrinol Metab, 2000. 85(12): p. 4434-40.
85. Oksnes, M., R. Ross, and K. Lovas, Optimal glucocorticoid replacement in adrenal insufficiency. Best Pract Res Clin Endocrinol Metab, 2015. 29(1): p. 3-15.
86. Bleicken, B., et al., Influence of hydrocortisone dosage scheme on health-related quality of life in patients with adrenal insufficiency. Clin Endocrinol (Oxf), 2010. 72(3): p. 297-304.
87. Riedel, M., et al., Quality of life in patients with Addison's disease: effects of different cortisol replacement modes. Exp Clin Endocrinol, 1993. 101(2): p. 106-11.
88. Debono, M., et al., Modified-release hydrocortisone to provide circadian cortisol profiles. J Clin Endocrinol Metab, 2009. 94(5): p. 1548-54.
89. Dallman, M.F., et al., Bottomed out: metabolic significance of the circadian trough in glucocorticoid concentrations. Int J Obes Relat Metab Disord, 2000. 24 Suppl 2: p. S40-6.
90. Ragnarsson, O., et al., The relationship between glucocorticoid replacement and quality of life in 2737 hypopituitary patients. Eur J Endocrinol, 2014. 171(5): p. 571-9.
91. Hahner, S., et al., Impaired subjective health status in 256 patients with adrenal insufficiency on standard therapy based on cross-sectional analysis. J Clin Endocrinol Metab, 2007. 92(10): p. 3912-22.
92. Smith, S.J., et al., Evidence that patients with Addison's disease are undertreated with fludrocortisone. Lancet, 1984. 1(8367): p. 11-4.
93. Staufenbiel, S.M., et al., Increased Hair Cortisol Concentrations and BMI in Patients With Pituitary-Adrenal Disease on Hydrocortisone Replacement. J Clin Endocrinol Metab, 2015. 100(6): p. 2456-62.
94. Filipsson, H., et al., The impact of glucocorticoid replacement regimens on metabolic outcome and comorbidity in hypopituitary patients. J Clin Endocrinol Metab, 2006. 91(10): p. 3954-61.
95. Bensing, S., et al., Increased death risk and altered cancer incidence pattern in patients with isolated or combined autoimmune primary adrenocortical insufficiency. Clin Endocrinol (Oxf), 2008. 69(5): p. 697-704.
96. Wolff, A.S., et al., Autoimmune polyendocrine syndrome type 1 in Norway: phenotypic variation, autoantibodies, and novel mutations in the autoimmune regulator gene. J Clin Endocrinol Metab, 2007. 92(2): p. 595-603.
97. Oftedal, B.E., et al., Dominant Mutations in the Autoimmune Regulator AIRE Are Associated with Common Organ-Specific Autoimmune Diseases. Immunity, 2015. 42(6): p. 1185-96.
98. Fassnacht, M. and B. Allolio, Clinical management of adrenocortical carcinoma. Best Pract Res Clin Endocrinol Metab, 2009. 23(2): p. 273-89.
44
99. Kebebew, E., et al., Extent of disease at presentation and outcome for adrenocortical carcinoma: have we made progress? World J Surg, 2006. 30(5): p. 872-8.
100. Kerkhofs, T.M., et al., Adrenocortical carcinoma: a population-based study on incidence and survival in the Netherlands since 1993. Eur J Cancer, 2013. 49(11): p. 2579-86.
101. Restifo, N.P., M.E. Dudley, and S.A. Rosenberg, Adoptive immunotherapy for cancer: harnessing the T cell response. Nat Rev Immunol, 2012. 12(4): p. 269-81.
102. June, C.H., Adoptive T cell therapy for cancer in the clinic. J Clin Invest, 2007. 117(6): p. 1466-76.
Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1154
Editor: The Dean of the Faculty of Medicine
A doctoral dissertation from the Faculty of Medicine, UppsalaUniversity, is usually a summary of a number of papers. A fewcopies of the complete dissertation are kept at major Swedishresearch libraries, while the summary alone is distributedinternationally through the series Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty ofMedicine. (Prior to January, 2005, the series was publishedunder the title “Comprehensive Summaries of UppsalaDissertations from the Faculty of Medicine”.)
Distribution: publications.uu.seurn:nbn:se:uu:diva-265276
ACTAUNIVERSITATIS
UPSALIENSISUPPSALA
2015