The role of intracellular thyroid hormone metabolism in innate immune cells van der Spek, A.H.

Transcription

1 UvA-DARE (Digital Academic Repository) The role of intracellular thyroid hormone metabolism in innate immune cells van der Spek, A.H. Link to publication Citation for published version (APA): van der Spek, A. H. (2018). The role of intracellular thyroid hormone metabolism in innate immune cells General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible. UvA-DARE is a service provided by the library of the University of Amsterdam ( Download date: 14 Aug 2018

2 GENERAL INTRODUCTION THYROID HORMONE METABOLISM AND INNATE IMMUNE CELLS Adapted from: A.H. van der Spek, E. Fliers, A. Boelen. The Classic Pathways of Thyroid Hormone Metabolism, Molecular and Cellular Endocrinology, 2017 Dec 15;458: A.H. van der Spek, E. Fliers, A. Boelen. Thyroid hormone metabolism and innate immune cells, Journal of Endocrinology, 2017 Feb;232(2):R67-R81

3 Chapter 1 Thyroid hormone metabolism Thyroid hormone production Thyroid hormones (TH) are crucial for growth and development and play an important role in energy homeostasis (Brent, 2012). TH is produced by the thyroid gland and then secreted into the bloodstream. The amount of TH in the blood is tightly regulated by the hypothalamus-pituitary-thyroid (HPT) axis. This is a classic example of an endocrine negative feedback loop which allows the brain to continuously sense the concentration of TH in the circulation and subsequently adapt the production of TH by the thyroid via the hypothalamus and the pituitary (Figure 1.1). Hypophysiotropic neurons within the paraventricular nucleus (PVN) of the hypothalamus produce thyrotropin-releasing hormone (TRH) which in turn stimulates the thyrotrope cells of the anterior pituitary to synthesize and secrete thyroid stimulating hormone (TSH) (Harris et al., 1978). TSH then stimulates the thyroid gland to produce TH in the form of thyroxine (T4) and triiodothyronine (T3) (Miot et al., 2015). Both TRH and TSH are negatively regulated by T3. In humans approximately 80% of the amount of TH secreted by the thyroid is in the form of T4, while 20% is the active form T3 (Maia et al., 2005). In rodents approximately 50% of the TH released by the thyroid is T4 and 50% is T3 (Chanoine et al., 1993). T4 functions as a prohormone and must be peripherally converted into T3 in order to become biologically active. This conversion occurs at the cellular and tissue level, enabling the local regulation of TH bioavailability. Figure 1.1 The hypothalamus-pituitarythyroid axis 12

4 General Introduction Thyroid hormone transport TH is actively transported into the cell by TH transporters (Figure 1.2). There are several families of TH transporters including organic anion transporter polypeptides (OATP), monocarboxylate transporters (MCT) and large neutral amino acid transporters (LAT) (Bernal et al., 2015, Visser, 2016). Of these transporters, MCT8 is the only one to transport TH exclusively. The other transporters are also capable of transporting additional substances including steroids and amino acids (Bernal et al., 2015). Studies in transgenic mouse models and observations in patients with pathogenic mutations in TH transporters indicate that MCT8, MCT10 and OATP1C1 are the main transporters that have been identified to date with (patho)physiological importance to TH transport in vivo (Bernal et al., 2015, Visser, 2016). MCT8 preferentially transports T4 whereas MCT10 preferentially transports T3 (Visser, 2016). OATP1C1 transports T3, T4 and rt3 with high specificity, and its affinity is the highest for T4 and rt3 (Bernal et al., 2015, Pizzagalli et al., 2002, Visser, 2016). Transporter expression is cell type specific and differences in distribution have been observed between humans and rodents (Bernal et al., 2015). 1 Figure 1.2: Intracellular TH metabolism D1: type 1 deiodinase; D2: type 2 deiodinase; D3: type 3 deiodinase; T3: triiodothyronine; T4: thyroxine; rt3: reverse triiodothyronine; T2: diiodothyronine; TRα: thyroid hormone receptor α; TRβ: thyroid hormone receptor β Deiodination After being transported into the cell, TH is metabolized by the iodothyronine deiodinases (Figure 1.2). The deiodinases are a family of enzymes that are capable of removing an iodine atom from the inner or outer ring of a TH isoform (Bianco and Kim, 2006, Visser and Peeters, 2012). Outer ring deiodination of T4 results in the activation of the hormone by generation of T3. Inner ring deiodination of T4 results in the formation of reverse T3 (rt3) (Figure 1.3). Although rt3 cannot bind the thyroid hormone 13

5 Chapter 1 receptor (TR) and was previously thought of as an inactive TH metabolite, it is now known to also have a variety of non genomic effects on for example actin remodeling and brain development (Farwell et al., 2006, Farwell et al., 2005, Leonard and Farwell, 1997) (see Davis et al (Davis et al., 2016) for a recent review). T3 can be inactivated by inner ring deiodination resulting in the formation of 3,3 -diiodothyronine (3, 3 -T2), a metabolite that can also be produced by outer ring deiodination of rt3 (Figure 1.1). Deiodinases are differentially expressed between various cell and tissue types (Gereben et al., 2008). Besides deiodination, there are other minor pathways of TH metabolism including sulfation, glucuronidation and ether link cleavage (Visser and Peeters, 2012, Wu et al., 2005, van der Spek et al., 2017a). Figure 1.3: Deiodination of thyroid hormone isoforms IRD: inner ring deiodination, ORD: outer ring deiodination Deiodinase type 1 Deiodinase type 1 (D1) is localized in the plasma membrane and highly expressed in liver, kidney, thyroid and pituitary (Bianco and Kim, 2006). D1 is able to deiodinate both the inner and the outer ring of T4, however its preferred substrate is not T4, but the inactive TH metabolites rt3, sulfated T3 (T3S) and sulphated T4 (T4S) (Moreno et al., 1994). Therefore, the primary role of D1 in vivo is to clear inactivated TH. Dio1 (the gene that encodes D1) expression is positively regulated by T3 via two TH responsive elements (TRE) in the D1 gene promoter (Jakobs et al., 1997, Toyoda et al., 1995). Deiodinase type 2 In contrast to D1, deiodinase type 2 (D2) is localized in the endoplasmic reticulum. D2 is expressed in 14

6 General Introduction many brain areas, pituitary, brown adipose tissue, placenta and although at remarkably low levelsskeletal muscle (Bianco and Kim, 2006, Gereben et al., 2008). In contrast to D1, D2 is involved in outer ring deiodination exclusively and the enzyme is thought to play a major role in local T3 production. The preferred substrate for D2 is T4 and, to a slightly lesser extent, rt3. D2 is regulated by THs both pre- and post-transcriptionally as T3 downregulates Dio2 (the gene that encodes D2) mrna expression (Burmeister et al., 1997), while T4 as well as rt3 (the substrates of D2) increase D2 ubiquitination and subsequent proteasomal degradation, resulting in decreased D2 activity. D2 protein half-life is very short (approximately 40 minutes) due to this ubiquitination process (Gereben et al., 2008). 1 Deiodinase type 3 Deiodinase type 3 (D3) is viewed as the TH inactivating enzyme, as it can only catalyze the inner ring deiodination of T4 and T3. Like D1, D3 is present in the plasma membrane (Baqui et al., 2003), although differences in the subcellular location of D3 have been described between cell types (Jo et al., 2012, Kallo et al., 2012, Alkemade et al., 2005). Dio3 (the gene that encodes D3) is highly expressed in placenta and plays an important role during embryonic development by regulating TH levels in the foetus. D3 is also expressed in the brain but in adult, healthy tissues expression levels are generally very low (Bianco and Kim, 2006, Gereben et al., 2008). D3 levels have been found to increase in various pathophysiological conditions. D3 is induced during critical illness (Peeters et al., 2003) and during hypoxia in various models of cardiovascular and neurological ischemia (Janssen et al., 2013, Simonides et al., 2008, Jo et al., 2012, Olivares et al., 2007). D3 induction during hypoxia is thought to be a protective mechanism as the resulting intracellular hypothyroidism presumably reduces cellular metabolism thus decreasing ischemic damage. Thyroid hormone receptors The classical pathway through which TH exerts its biological effects is by binding to its nuclear TH receptors (TRs). Upon binding of T3, these TRs are capable of directly initiating or inhibiting gene transcription (Mullur et al., 2014, Brent, 2012) (Figure 1.2). Thyroid hormone receptors (TRs) are encoded by the thyroid hormone receptor α (Thra) and thyroid hormone receptor β (Thrb) genes. These two genes can generate several TR isoforms by alternative splicing of their transcription products (Cheng et al., 2010). These isoforms are differentially expressed in a tissue and cell type specific manner (Brent, 2012). There are three TRα isoforms of which only TRα1 is a classic ligand binding receptor capable of binding T3. TRα2 and TRα3 cannot bind T3 (Cheng et al., 2010). TRα1 is widely expressed in cardiac and skeletal muscle, the central nervous system and bone (Cheng et al., 2010, Brent, 2012). The two main TRβ isoforms capable of binding T3 are TRβ1 and TRβ2 (Cheng et al., 2010). TRβ1 is mainly present in the brain, liver, kidney, heart and thyroid (Cheng et al., 2010). TRβ2 is predominantly expressed in the hypothalamus, pituitary, retina and inner ear (Cheng et al., 2010). The identification of patients with inactivating mutations of either TRα or TRβ has further illustrated the tissue distribution pattern of these receptors. Resistance to TH due to a TRβ mutation (RTHβ) results in severely abnormal plasma TH levels characterized by elevated TH and unsuppressed TSH due to the lack of functional TRβ within the HPT-axis (Yen, 2003). In contrast, RTHα patients display signs of hypothyroidism at the tissue 15

7 Chapter 1 level such as growth retardation, neurological abnormalities and constipation despite having only mild abnormalities in circulating TH levels (Moran and Chatterjee, 2015). There is increasing evidence that THs also act via non genomic pathways (Davis et al., 2016). The pathways involved in non-genomic TH actions are initiated by binding of TH to another receptor than the intracellular TRs, for example to the plasma membrane receptor integrin αvβ3 (Davis et al., 2016). The classic pathways of TH action and the rapid non genomic pathways activated by TH are not completely independent from each other as rapid non genomic actions of TH can affect intracellular TRs and even require TRs in certain cell types (Davis et al., 2016, Flamant, 2016). Innate Immune Cells The innate immune system is responsible for the host defence against invading pathogens. The cells of the innate immune system identify microbes, initiate an inflammatory response and can either directly phagocytose and kill pathogens or recruit other innate or adaptive immune cells to the site of infection. Innate immune cells are derived from hematopoietic stem cells in the bone marrow. These cells can be mobilized from the blood or bone marrow upon infection. Alternatively, innate immune cells can travel from the bone marrow to the tissue and patrol there for invading pathogens, and these cells are known as tissue resident cells Neutrophils Neutrophils are the first cells to be recruited to the site of inflammation and are the most abundant type of blood leukocyte, comprising 50-75% of circulating leukocytes in humans (Borregaard, 2010, Bardoel et al., 2014, Kolaczkowska and Kubes, 2013). Circulating neutrophils are generated in the bone marrow by hematopoietic stem cells and are short lived cells with an estimated life span of hours to days (Borregaard, 2010, Kolaczkowska and Kubes, 2013). Upon inflammation and infection, neutrophils from the circulation are recruited to the site of inflammation. Inflammatory mediators are recognised by neutrophils, after which they adhere to the vascular endothelium close to the site of infection before transmigrating into the extravascular tissue. Extravasated neutrophils then migrate to the place of inflammation where they can then kill invading pathogens and secrete inflammatory mediators further stimulating the immune response and recruiting other innate and adaptive immune cells (Kolaczkowska and Kubes, 2013). Neutrophils are highly specialized cells that have multiple microbial killing mechanisms at their disposal. The three main killing mechanisms utilized by neutrophils are degranulation, the production of reactive oxygen species and the generation of neutrophil extracellular traps (Bardoel et al., 2014, Kolaczkowska and Kubes, 2013) (Figure 1.4). Upon phagocytosis of a pathogen, neutrophils can release various bactericidal elements into the phagosome. One of these elements are antimicrobial proteins and enzymes that are formed sequentially during neutrophil development and stored in intracellular granules (Borregaard, 2010, Borregaard and Cowland, 1997). Upon phagocytosis these granules can fuse with 16

8 General Introduction the phagosome or the plasma membrane, releasing their contents in a process known as degranulation (Borregaard, 2010) (Figure 1.4). Neutrophils are also capable of generating reactive oxygen species (ROS) in the phagosome using the nicotinamide adenine dinucleotide phosphate (NADPH) oxidase system (Figure 1.4). ROS can also be released into the extracellular space. An important extracellular killing mechanism are neutrophil extracellular traps (NETs) (Brinkmann et al., 2004). NETs are composed of neutrophil chromatin to which antimicrobial proteins and ROS are bound (Brinkmann et al., 2004, Kolaczkowska and Kubes, 2013). The release of these NETs enables neutrophils to effectively trap and kill extracellular bacteria, but also eventually results in the death of the neutrophil (Brinkmann et al., 2004). 1 Figure 1.4: Neutrophil bacterial killing mechanisms; After encountering a pathogen, neutrophils can phagocytose the microbe and release antibacterial proteins or reactive oxygen species (ROS) into the phagosome or extracellular space. Neutrophils can also generate neutrophil extracellular traps (NETs) which consist of neutrophil DNA, ROS and antibacterial proteins. Monocytes and macrophages Monocytes and macrophages are mononuclear phagocytic cells. Monocytes are continuously generated in the bone marrow by hematopoiesis and released into the circulation where they constitute 10% of circulating human leukocytes. There is also a considerable monocyte reservoir in the spleen and lungs that can be mobilized on demand (Ginhoux and Jung, 2014). Circulating monocytes can extravasate to tissues both during the steady state and during inflammation where they can differentiate into macrophages or dendritic cells (Shi and Pamer, 2011). An alternative subset of macrophages are the tissue resident macrophages. Until recently, these were thought to be continuously replenished from the circulating monocyte pool. Tissue resident macrophages are now known to be derived from embryonic precursors that colonize the tissues prenatally (Mass et al., 2016). These cells, which include Kupffer cells and microglia, are also able to maintain their populations in adult tissues due to local cell proliferation independently of circulating monocytes (Ginhoux and Jung, 2014, Hashimoto et al., 2013, Mass et al., 2016). The various tissue resident macrophages comprise distinct cell populations whose phenotype differs strongly between tissues (Murray and Wynn, 2011). 17

9 Chapter 1 After entering the tissue, macrophages can change their phenotype dependent on various stimuli, allowing them to adapt to a wide subset of roles. This process is known as polarization. Polarized macrophages are generally classified into M1 or pro-inflammatory macrophages, and M2 or alternatively activated macrophages which are a heterogeneous group of cells that have a more anti-inflammatory profile (Murray and Wynn, 2011, Murray et al., 2014). Pro-inflammatory macrophages are important in antimicrobial defence and the recruitment of neutrophils and T cells to the inflamed tissue (Murray and Wynn, 2011). They are also capable of antigen presentation and can elicit a T-cell response (Hume, 2008). Polarization of macrophages into a pro-inflammatory phenotype is accompanied by changes in cellular metabolism, shifting towards enhanced glycolysis (Freemerman et al., 2014, Galvan-Pena and O Neill, 2014, Zhu et al., 2015). Essential components of adequate pro-inflammatory macrophage function are phagocytosis, the generation of ROS by NADPH oxidase and the generation of reactive nitrogen species (RNS) which is mediated by inducible nitric oxide synthase (inos) (Weiss and Schaible, 2015). M2 macrophages are tolerogenic and immunomodulatory cells that are involved in wound healing and tissue remodelling (Murray and Wynn, 2011). This is also accompanied by metabolic changes resulting in enhanced fatty acid oxidation and mitochondrial oxidative phosphorylation (Galvan-Pena and O Neill, 2014, Vats et al., 2006, Zhu et al., 2015). More recent data suggests that macrophage polarization is not as clear cut as these two phenotypes and represents more of a spectrum ranging from pro- to antiinflammatory (Hume, 2015, Murray et al., 2014). Thyroid hormone metabolism in innate immune cells Thyroid hormone metabolism in neutrophils It has long been known that activated neutrophils are capable of deiodinating both T3 and T4 (Woeber, 1978, Woeber et al., 1972, Woeber and Ingbar, 1973, Woeber, 1971, Klebanoff and Green, 1973). Research from the seventies already found that phagocytosing human neutrophils can generate both T3 and r T3 from T4 and that this deiodinating activity was mainly present in the granule fraction of the cells (Woeber, 1978, Woeber, 1976a). Neutrophils were also shown to contain saturable nuclear binding sites for T3 (Woeber, 1977). More recently, our lab found that type 3 deiodinase (D3), the TH inactivating enzyme, is present in infiltrating murine neutrophils (Boelen et al., 2008, Boelen et al., 2005). Murine neutrophils were also found to express the TH transporter MCT8 (Boelen et al., 2005). Research from the seventies using radioactively labelled TH demonstrated that THs are drawn to the site of bacterial infection (Adelberg et al., 1971). Activated phagocytosing neutrophils are capable of cleavage of thyroxine-binding globuline (TBG), thus increasing the amount of extracellularly available T4 (Jirasakuldech et al., 2000). In addition, phagocytosing neutrophils also metabolize significant amounts of TH (Woeber, 1978, Woeber et al., 1972, Woeber and Ingbar, 1973, Woeber, 1971, Klebanoff and Green, 1973). This suggests that TH metabolism plays an important role in infiltrating neutrophils during infection. Metabolism of TH by phagocytosing neutrophils results in the production of inorganic iodide, suggesting the involvement of the deiodinase enzymes (Klebanoff and Green, 1973, Woeber et al., 18

10 General Introduction 1972, Woeber and Ingbar, 1973). However, some authors have found that phagocytosing neutrophils are capable of ether link cleavage of T4 which results in the formation of diiodotyrosine (DIT) (Burger et al., 1983). Neutrophils from patients with chronic granulomatous disease, characterized by defective neutrophil NAPDH oxidase activity and impaired ROS generation, are significantly less effective at degrading TH (Klebanoff and Green, 1973, Woeber and Ingbar, 1973, Burger et al., 1983). This suggests that degradation of TH requires the intracellular formation of ROS. Although isolated MPO is capable of degrading TH in vitro, neutrophils from MPO-deficient patients degrade TH to the same degree as controls suggesting that the degradation of TH by leukocytes is not MPO dependent in vivo (Klebanoff and Green, 1973, Woeber and Ingbar, 1973, Burger et al., 1983). 1 Type 3 deiodinase (D3) is expressed in infiltrating murine neutrophils during both bacterial infection and sterile inflammation due to a turpentine injection which forms a subcutaneous abscess (Boelen et al., 2008, Boelen et al., 2005). Mice that lack D3 have impaired bacterial killing upon infection with Streptococcus pneumoniae (Boelen et al., 2009). Together these data suggest that D3 is important for neutrophil function during infection and inflammation. The mechanism behind this is currently unknown. It has previously been suggested that the iodide produced by D3 could be used by the MPO system together with H 2 O 2 to generate hypoiodite, a toxic compound that is capable of killing bacteria (Boelen et al., 2011, Klebanoff, 1967). Thyroid hormone metabolism in monocytes and macrophages Macrophages contain several essential elements of intracellular TH metabolism. Both macrophage cell lines and human and murine microglia contain TH transporters. Macrophage cell lines predominantly express MCT10 and to a lesser extent MCT8 (Kwakkel et al., 2014). Microglia, the resident macrophages of the brain, contain the TH transporters LAT2, MCT10 and OATP4a1 (Braun et al., 2011, Wirth et al., 2009). Macrophages were also found to express D2 (Kwakkel et al., 2014), TRα1 and possibly also TRβ although authors have reported conflicting results (Billon et al., 2014, Kwakkel et al., 2014, Lourbopoulos et al., 2014, Perrotta et al., 2014). Thyroid hormone levels appear to play an important role in macrophage function. TH administration increases inos expression, nitrite production and in vitro bacterial killing in both a human and a mouse macrophage cell line while treatment with TH increased survival after meningococcal infection in mice (Chen et al., 2012). These effects were found to be partly mediated via binding of TH to integrin αvβ3 on the extracellular surface of the cell resulting in the rapid activation of the PI3K and ERK1/2 signalling pathways (Chen et al., 2012). Besides the extracellular binding of TH, regulation of intracellular TH levels was also recently shown to play an essential role in the pro-inflammatory response of macrophages (Kwakkel et al., 2014). D2, which converts T4 to T3 thereby regulating intracellular TH bioavailability, is induced in macrophages stimulated with bacterial endotoxin (lipopolysaccharide; LPS) together with TRα1 and MCT10, indicating a shift towards increased TH action during inflammation (Kwakkel et al., 2014). Furthermore, D2 knockdown resulted in impaired macrophage phagocytosis and blunted cytokine response to LPS 19

11 Chapter 1 stimulation (Kwakkel et al., 2014). These effects appear to be partly mediated via genomic pathways as knockout of TRα, which is the predominant TR isoform in macrophages, also results in aberrant macrophage function (Kwakkel et al., 2014, Billon et al., 2014). Macrophages from TRα knockout mice have impaired cholesterol efflux during atherosclerosis resulting in earlier plaque formation (Billon et al., 2014). Furthermore, macrophages that lack TRα demonstrate low grade inflammation at baseline compared to controls, indicating an anti-inflammatory role for TRα (Billon et al., 2014). This suggests that attenuation of the rapid pro-inflammatory response by increased intracellular TH levels could be mediated via TRα. Research questions and thesis outline In this thesis we aimed to determine the role of intracellular TH metabolism in the function of neutrophils and macrophages. In Chapter 2 we determined components of intracellular TH metabolism in murine neutrophils, macrophages and bone marrow precursor cells. As D3 was expected to play an important role in neutrophil function based on several mouse models, we studied the presence and subcellular distribution of D3 in primary human neutrophils in Chapter 3. In Chapter 4, we aimed to establish an in vitro model using a neutrophil-like cell line to enable more mechanistic studies into the role of D3 in neutrophils. In Chapter 5, we aimed to determine whether D3 plays a functional role in neutrophil bacterial killing by using in vitro (human/mouse) and in vivo (zebrafish) models. In macrophages, both D2 and TRα have been suggested to affect cell function. Therefore, in Chapter 6 we studied neutrophil and macrophage function in a patient with an inactivating mutation of TRα. Finally, in Chapter 7 we studied the effects of changes in D2 and TRα on pro-inflammatory macrophage function. Finally, we discuss the overall results in a broader context and identify future perspectives in Chapter 8. 20