The role of peroxisomes in malnutrition and metabolic disease José Manuel Horcas Nieto In vitro and in silico models
The role of peroxisomes in malnutrition and metabolic disease In vitro and in silico models José Manuel Horcas Nieto
This work was supported by the European Union's Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 812968 (PeRIco), the Canadian Institutes of Health Research (156307), the Stichting De Cock-Hadders and Vrienden Beatrix Kinderziekenhuis. The research presented in this thesis was conducted at the Department of Pediatrics, Center for Liver and Metabolic Diseases, University Medical Center Groningen, the Netherlands. The printing of this thesis was financially supported by University of Groningen and the Graduate School of Medical Science of the University Medical Center Groningen. Printing: Ridderprint, ridderprint.nl Layout and design: Ridderprint | www.ridderprint.com © Copyright 2024 Jose Manuel Horcas Nieto, Groningen, The Netherlands. All rights reserved. No part of this thesis may be reproduced, distributed, sorted in a retrieval system, or transmitted in any form o by any means without prior permission of the author or, where applicable, the publisher holding the copyright on the published articles.
The role of peroxisomes in malnutrition and metabolic disease In vitro and in silico models PhD thesis to obtain the degree of PhD of the University of Groningen on the authority of the Rector Magnificus and in accordance with the decision by the College of Deans. This thesis will be defended in public on Wednesday 8 May at 14.30 hours by Jose Manuel Horcas Nieto born on 10 April 1993 Córdoba
Supervisor(s) Prof. B.M. Bakker Co-supervisor(s) Dr. R.H.J. Bandsma Assessment Committee Prof. R.J.A. Wanders Prof. A. M. Dolga Prof. K.N. Faber
TABLE OF CONTENTS Chapter 1: General Introduction 7 Chapter 2: Organoids as a model to study intestinal and liver dysfunction in severe malnutrition 33 Chapter 3: Towards Automatization of Organoid Analysis: A Deep Learning Approach to Localize and Quantify Organoids Images 73 Chapter 4: Docosahexaenoic acid prevents peroxisomal and mitochondrial protein loss in a murine hepatic organoid model of severe malnutrition 89 Chapter 5: Establishing a peroxisomal β-oxidation computational kinetic model to understand the effects of amino-acid restriction on peroxisomal β-oxidation 129 Chapter 6: iPSC-derived liver organoids as a tool to study Medium Chain Acyl-CoA Dehydrogenase deficiency 167 Chapter 7: General Discussion 201 Appendices Summary 218 Samenvatting 222 Resumen para público no científico 227 Acknowledgments 232 Curriculum vitae 238 List of publications 240
7 1 General Introduction Chapter General Introduction
8 Chapter 1 FATTY ACID METABOLISM Fatty acids are important molecules with multiple roles, such as structural components of cell membranes, nutrients and signaling molecules1. They can be stored as triglycerides to be used during extended periods of fasting, illness or exercise. Fatty acid oxidation (FAO) provides a large amount of the energy requirements during periods when glucose and gluconeogenic precursors are depleted. FAO is a key metabolic process for human physiology, and its importance is highlighted by the number of inherited disorders caused by defects in the pathway2. While the majority of fatty acids are oxidized via mitochondrial β-oxidation, peroxisomes also play a key role in FAO of different substrates. Interestingly, these two organelles are able to communicate and interact with each other in the oxidation of fatty acids. Organelles involved in metabolism of fatty acids Mitochondria Mitochondria are subcellular organelles in charge of the bulk of fatty-acid oxidation. These double-membrane organelles are in charge of the production of the majority of the ATP in the cell through oxidative phosphorylation3. Amongst many of the functions of mitochondria I could highlight energy metabolism, redox metabolism, calcium regulation, and signaling 4. Regarding fatty acid metabolism, mitochondria are in charge of oxidizing long- to short-chain fatty acids to produce ATP under catabolic conditions. To do so, mitochondria can perform β-oxidation of fatty acids, a cyclic process that leads to the production of acetyl-CoA and NADH and FADH2 2. These products are further metabolized in the tricarboxylic acid (TCA) cycle and oxidative phosphorylation where ATP is formed. Mitochondria are highly dynamic organelles, which continuously experience processes of recycling and reshaping5,6. They are also able to respond to their environment regulating both their structure and function. Defects in mitochondrial metabolism, namely inherited defects, DNA damage during aging and environmental factors, such as diet can be detrimental for cell survival7. Peroxisomes Peroxisomes are small (0.1–1 µm), single-membrane subcellular organelles found in most eukaryotic cells. These organelles exert a wide range of metabolic
9 1 General Introduction functions, which vary between different organisms and even between different organs in a single organism8. Some of the metabolic functions of peroxisomes include lipid metabolism, hydrogen peroxide metabolism, synthesis of ether lipids and bile acids, glyoxylate detoxification, and in trypanosomatids even glycolysis89. Impairments in peroxisomal function can lead to severe diseases, highlighting the importance of these organelles. Peroxisomal disorders can be divided in peroxisomal biogenesis disorders (PBDs) and single-enzyme deficiencies. Peroxisomal biogenesis disorders are autosomal recessive inherited disorders caused by a defect in any of the PEX genes8. These genes encode for proteins (peroxins) taking part in peroxisomal biogenesis. Peroxisomal biogenesis disorders affect peroxins involved in either the import of peroxisomal proteins10,11,in peroxisomal division12–14or de novo formation15,16. These disorders lead to multiple metabolic dysfunctions. Single-enzyme deficiencies are caused by impairments or defects of any enzyme involved in specific peroxisomal metabolic pathways (e.g. ABCD1, ACOX1, AMACR, etc.). Unlike PBDs, singleenzyme deficiencies only affect specific metabolic functions. One key function of peroxisomes, and the main point of interest of this thesis, is fatty acid metabolism. Peroxisomes are able to metabolize several types of fatty acids, including very long chain fatty acids (VLCFA), branched chain fatty acids (BCFA) and even medium-chain fatty acids17. Moreover, peroxisomes are also capable of oxidizing mono- and poly-unsaturated fatty acids18. To do so, peroxisomes can perform both α- and β-oxidation with a very broad range of substrates. Although peroxisomal β -oxidation is similar to the mitochondrial pathway, peroxisomes are equipped with their own set of enzymes in charge of metabolizing both straight-chain acyl-CoA to Acetyl-CoA as well as 2-methylbranched-chain acyl-CoAs, into propionyl-CoA and Acetyl-CoA18. The interplay between peroxisomes and mitochondria In recent times, it has become clear that peroxisomes and mitochondria display a functional interplay, key for different metabolic processes19–21. Reactive oxygen species (ROS) are molecules formed during different cellular processes that contribute to aging processes. While essential for cell signaling, when accumulated ROS can cause oxidative damage22. As an example of the interaction between peroxisomes and mitochondria, we could highlight their combined role in ROS homeostasis 23. Another relevant example in the context of this thesis is the oxidation of fatty acids in order to maintain lipid homeostasis. While the mechanisms of interaction between these two
10 Chapter 1 organelles have not been deeply characterized they involve physical contact sites, diffusion processes, and vesicular transport24–26. Given the close interplay between the two organelles, it has been suggested that alterations in one of them (either in biogenesis, proliferation or metabolism) might potentially have an effect on the other27. In the case of peroxisomes, several studies have reported that diseases affecting peroxisomal fatty acid metabolism28, peroxisomal biogenesis29 and peroxisomal redox activity30 also had a negative effect on mitochondrial health. This phenomenon was also observed in hepatocytes of a liver-specific knockdown of PEX5. L-Pex5−/− mice showed alteration of the mitochondrial inner membrane, increased oxidative stress and depletion of mitochondrial DNA31. On the other hand, it remains unknown if primary mitochondrial diseases also cause peroxisomal dysfunction. These findings not only support the notion of a tight interplay between both organelles, but also point towards potential compensatory mechanisms between the two organelles. Given the importance of both mitochondria and peroxisomes in the metabolism of fatty acids, it is understandable that defects in either one of these organelles might have serious health repercussions. Disruptions in normal fatty acid metabolism caused by nutritional and genetic disorders Imbalances in different biochemical processes involved in fatty acid metabolism cause a vast range of human diseases. The origin of these imbalances can be either genetic (inborn errors of metabolism) or environmental factors and contextual conditions (such as nutrition and/or malnutrition). Different nutritional stressors can have an important impact on the metabolism of macronutrients. There are multifactorial diseases which are caused by a combination of multiple genetic and dietary factors, of which none is in itself pathogenic. A common example of an acquired, multifactorial metabolic disease is NonAlcoholic Fatty Liver Disease (NAFLD), known as one of the most prevalent chronic liver diseases. NAFLD is associated with different risk factors such as diabetes type II, obesity and metabolic syndrome32,33. A contrasting example of another disease that leads to fatty liver associated with disturbances in fatty metabolism is undernutrition 34–38. Interestingly, illustrating the complexity of this topic, moderately-low protein diets without malnutrition, have been shown to increase lifespan and metabolic health in mice39,40.
11 1 General Introduction On the other end of the spectrum, there are also monogenic diseases where a single mutation can lead to disturbances in fatty acid metabolism. These genetic diseases, which lead to a deficiency of a single enzyme of any metabolic pathway related to the metabolism of carbohydrates, fats or protein are known as inborn errors of metabolism (IEM)41. Some examples of inborn errors of metabolism affecting mitochondrial β-oxidation are deficiencies in any of the dehydrogenases involved in the first step of the pathway (SCADD, MCADD and VLCADD)42. IEMs often present during childhood, and they can lead to the development of life-threating symptoms. Despite the fact that extensive research is currently being done, the fact remains that these diseases are commonly only managed palliatively as no cure is known for most of them. This thesis focuses on two diseases affecting lipid metabolism on both ends of the spectrum: one caused by an imbalance of the diet (severe malnutrition) and the other by a genetic mutation in the ACADM gene (MCADD). SEVERE MALNUTRITION The first disease central to the thesis is Severe malnutrition (SM). Malnutrition refers to an imbalance or deficit in the intake of energy and nutrients required to maintain homeostasis. Malnutrition can be caused by either a deficiency or an excess of nutrients, undernutrition and overnutrition respectively43. In this thesis, the term malnutrition is used to refer to undernutrition. Severe malnutrition is known as the most severe form of macronutrient deficiency. In the case of children, this can hamper optimal growth and development44. There are different types of malnutrition, namely stunting, wasting, underweight and deficiencies of vitamins and minerals. Stunting refers to children who are too short for their age. Wasting is used to describe children who are too thin for their height. Finally, the term undernutrition engulfs both stunting and wasting. Severe malnutrition is a highly challenging global health problem: in 2023 approximately 45 million children were wasted, of whom 13.7 million were severely wasted45. Severe malnutrition is common in low-income countries, where diets are often high in carbohydrates and contain low amounts of protein46. Children suffering from malnutrition present with a wide spectrum of symptoms affecting the liver and the intestine amongst other organs47. Severely malnourished children often suffer from intestinal dysfunction, including diarrhea and increased permeability of the intestinal barrier. These factors can contribute to increased risk of dehydration and sepsis respectively48.
12 Chapter 1 Additionally, the intestine of malnourished children has been described to be severely impacted not only in function but also in structure49. Regarding the effects of malnutrition in the liver, hepatic steatosis, hypoalbuminemia and hypoglycemia are some of the common problems observed in malnourished children50,51. Hepatic steatosis in malnourished children has been linked to the observed impaired lipid metabolism.52,53 Moreover, the number of peroxisomes was found to be substantially decreased in the liver of malnourished children54. Current approaches for the study of malnutrition mostly rely on the use of animal models, including primates55, rats56 and more frequently, mice57,58 on low protein diets (LPD). These models, have shed some light onto the pathophysiology of malnutrition in the liver and the intestine and are able to recapitulate the specific phenotypes of the disease (e.g., hepatic steatosis, barrier dysfunction, etc.). Some of these studies also revealed dysfunctional mitochondria in the liver and the intestine of both mice and rats 56,58 while peroxisomal loss in the liver was recapitulated in the rat model56, but not yet studied in mice. While these models are physiologically relevant, and allow for a whole-body characterization of the disease, they are also limited by blood and tissue sampling and require large numbers of animals for large screenings. These limitations emphasize the need for relevant in vitro malnutrition models to study organ-specific phenotypes as well as intercellular communication. The role of peroxisomes and mitochondria in severe malnutrition An interesting feature of peroxisomes is their plasticity and their ability to respond to different stimuli and adapt to the cellular needs59. This behavior goes in line with the response observed in malnourished children fed low protein diets. Peroxisomes have been reported to substantially decrease in number in the liver of malnourished children and rodents on low protein diets (Figure 1)54,56–58. Moreover, mitochondria were also found to be impaired, characterized by defects in complexes I and IV in vivo, and reduced hepatic ATP levels. Dysfunction in both organelles might explain the accumulation of fatty acids in the liver of malnourished children leading to hepatic steatosis52. In the case of the intestine of mice fed an LPD, mitochondrial abundance was reduced with a clear reduction in protein expression of complexes I and V from the electron transport chain (ETC) (Figure 1). Moreover, levels of reactive oxygen species were found to be increased. These results, together with decreased autophagy, were causally linked to increased permeability observed
13 1 General Introduction in the intestine of the animals60. Nothing has been published on the number and health of peroxisomes in the intestine of mice fed a low protein diet. Figure 1. The effects of low protein diets and low amino acid conditions in vivo. Low protein diets lead to a loss of peroxisomes and delayed loss of mitochondria in the liver. LPD leads to dysfunctional and less abundant mitochondria in the intestine. There is no information on the number of peroxisomes in the intestine of LPD rodents or children. Created with BioRender. com Interestingly, little is known about the actual mechanisms of peroxisomal loss in the liver of malnourished children and mice. Two main explanations have been suggested: a potential increase in autophagic degradation in order to maintain levels of precursors in the cell or a reduced biogenesis due to lack of amino acids56.
14 Chapter 1 Autophagic degradation and its role in malnutrition Both mitochondria and peroxisomes are commonly degraded via selective autophagy61,62. Autophagy is a degradation process based on the delivery of cargo, encapsulated in autophagosomes, to the lysosome, where different biological polymers are broken down by a combination of enzymes63. The autophagic process starts by the formation of a structure known as the phagophore. Once the membrane of the phagophore closes around its cargo, it forms the autophagosome. This then fuses with the lysosome forming the autolysosome in which the lysosome secrets a set of enzymes able to degrade proteins, lipids and carbohydrates64,65. Traditional autophagy markers, mentioned throughout this thesis, are LC3-I, LC3-II and p62. LC3-I is conjugated with phosphatidylethanolamine to form LC3-II which is then targeted to the autophagosomal membrane. Once the autophagosome fuses with the lysosome, LC3-II is degraded. Therefore, the LC3-II/LC3-I is commonly used to track the autolysosome formation. P62 is an autophagy adaptor protein which delivers polyubiquitinated cargo to the phagophore66. This process is regulated in response to different nutritional states such as amino acid starvation, a well-known autophagy activator67–69. Since most proteins and organelles are degraded via autophagy, regulation of this process grants autophagy the ability to maintain the levels of amino acids, especially during scarcity of these nutrients70. The decrease in the number of peroxisomes in the liver of rats on a low protein diet has been linked to autophagic degradation56. These results were in line with the induction of autophagy in vitro using amino acid deprivation67. Moreover, peroxisomes have been described to be highly sensitive to amino acid levels and have been reported to rapidly degrade via pexophagy, selective peroxisomal autophagy, under low-amino acid concentration conditions71. Another example of peroxisomes responding to nutrient deprivation is the induction of pexophagy in yeast under nitrogen starvation conditions72. A low-protein-diet study in rats showed induction of autophagy after 1 week, illustrated by elevated levels of LC3-II and reduced levels of p62. Interestingly, after 4 weeks of LPD the levels of p62 were elevated while LC3-II remained high. These results were interpreted as a potential block in autophagy after prolonged starvation56. Some other studies in mice showed that already two weeks of LPD led to a decreased autophagy activation in mice57,58. All these studies focused on the assessment of autophagy based on snapshots of LC3 and P62 at given times. However, it is important to take into consideration that autophagy is a dynamic process. While LC3-II levels are known to relate to the
15 1 General Introduction number of autophagic structures in the cells, they do not quantify the actual autophagic flux73. Very few papers have been published on the role of autophagy in the intestine of LPD-fed rodents. Only one study has reported that 2 weeks of low protein diet led to a decreased autophagy in the intestine of rodents, which was thought to be the cause of the observed accumulation of dysfunctional mitochondria60. Peroxisomal biogenesis and PPAR-α activation To date, no clear link between amino-acid restriction and decreased peroxisomal biogenesis has been made. One in vivo studied in malnourished rats showed no clear effect of LPD on the expression of different peroxins56. Peroxisome proliferator-activated receptor (PPAR)-α is a transcriptional factor belonging to the family of nuclear receptors. It is an important regulator of energy homeostasis74 and it controls the expression of genes involved in β-oxidation of fatty acids75,76. Moreover, PPAR- α activation is known to induce proliferation of peroxisomes77. Traditionally, PPAR-α activators, such as fibrates, have been used to treat dyslipidemia as well as to induce biogenesis of peroxisomes and mitochondria78. While no clear effect of malnutrition on peroxisomal and mitochondrial biogenesis has been described, the effect of these compounds has been previously studied in the context of malnutrition56,79. It is important to mention that some of these compounds have also been reported to cause side effects. For example, fenofibrate was able to reduce plasma triglycerides in PPAR-α null mice, as expected, but it also increased the levels of intrahepatic triglycerides80. Different natural ligands, such as eicosanoid derivatives and long chain polyunsaturated fatty acids (LCPUFA) have been reported to activate PPARα76,81,82. The ability of these natural compounds to activate PPAR-α, and thereby peroxisomal and mitochondrial biogenesis, opens the door to potential new therapies for malnutrition based on natural food supplements. In this line, several studies have already been published on the positive effects of the LCPUFA docosahexaenoic acid (DHA) on different aspects of malnutrition, such as antioxidants metabolism in rodents83 and regulation of cognition in malnourished infants84. Both of these studies highlighted the benefits of DHA supplementation in severe malnutrition.
16 Chapter 1 MEDIUM CHAIN ACYL-COA DEHYDROGENASE DEFICIENCY The second disease studied in this thesis is a monogenic disease known as Medium Chain Acyl-CoA Dehydrogenase Deficiency (MCADD). Medium-chain Acyl-CoA dehydrogenase (MCAD) is a flavoprotein containing a flavin adenine nucleotide (FAD). It catalyzes the oxidation of an acyl-CoA with the simultaneous reduction of the flavin, which then needs to be reoxidized85. MCAD is involved in the oxidation of medium chain fatty acids in the mitochondrial β-oxidation. Prior to being metabolized, fatty acids need to be activated into their CoA ester form. The acyl-CoA esters can then be oxidized to acetyl-CoA in order to produce NADH and FADH2. This pathway is of special importance during fasting or situations of high energy demand (cold exposure, infection, etc.) when gluconeogenic precursors are low and we depend on our fat reserves. Medium Chain Acyl-CoA Dehydrogenase Deficiency is the most common fatty acid oxidation disorder with a prevalence of 1:8000 in the Netherlands86. Biochemically, MCADD is characterized by the accumulation of medium-chain monocarboxylic acids and acyl-carnitines. Fatty acids of more than 12 carbons (C12) are often considered to be long while those between 10 and 6 carbons (C10-C6) are referred to as medium-chain fatty acids. Given the lack of efficient oxidation of medium-chain fatty acids, MCADD is also characterized by a high ratio C8 over C10 acylcarnitines87–89. Clinically, MCADD patients are at risk of suffering hypoketotic hypoglycemia (low levels of blood glucose and ketone bodies) which can be fatal2. Interestingly, heterogeneity in symptomatology in MCADD patients is high. While some patients present with severe symptoms, other patients never develop any symptoms, even when carrying the same mutation and without surveillance90–92. While MCADD patients present as healthy under most conditions, they are insufficiently able to oxidize medium-chain fatty acids when they rely on fatty acid reserves such as fasting, illness or long bouts of exercise93. Current methods for the study of MCADD include animal models (rodents)94,95 and artificial KO cell lines. While these models have shed some light onto the pathophysiology of MCADD, both have their own limitations. Although animal models are wholistic systems that allow for a physiologically accurate picture, rodents express a long-chain acyl-CoA dehydrogenase isoform in charge of oxidizing long-chain acyl-CoAs96. LCAD shares substrate specificity with MCAD which might alleviate the phenotype. On the other hand, KO human cell lines allow us to study the disease without expression of LCAD but are limited in terms of cell functionality and lack interaction with other organs. Moreover,
17 1 General Introduction none of these systems allow researchers to address the issue of patient heterogeneity, emphasizing the need for new experimental models to study MCADD in patient derived tissue. The role of peroxisomes in MCAD deficiency As mentioned above, MCAD deficiency is characterized by the accumulation of substrates that are normally metabolized by the medium-chain dehydrogenase. Under non-challenging conditions other dehydrogenases are able to prevent severe phenotypes. In contrast, catabolic conditions such as fasting and recurring illness can trigger accumulation of medium-chain fatty acids in MCAD deficient patients. Peroxisomes have been reported to metabolize a wide range of substrates including medium-chain acyl-CoAs when mitochondrial import of fatty acids is inhibited97. Interestingly, peroxisomes are equipped with all the enzymes required for the activation, import and oxidation of medium-chain fatty acids. A study reported that medium-chain acyl-CoAs can enter peroxisomes via ABCD3 (PMP70) prior to being oxidized by the peroxisomal β-oxidation. These results highlight peroxisomes as a potential route to metabolize medium-chain fatty acids in MCAD deficiency in order to prevent metabolite accumulation. Yet, no studies have reported on the role of peroxisomes in any of the mitochondrial dehydrogenase deficiencies (SCADD, MCADD or VLCADD). NEW EXPERIMENTAL AND COMPUTATIONAL MODELS TO STUDY SEVERE MALNUTRITION AND MCADD IN VITRO AND IN SILICO I have already introduced some of the more traditional models for the characterization and study of both severe malnutrition and MCADD. However, it is important to realize that all models are flawed and present both advantages and limitations. In the context of this thesis, I introduce two new tools for the study of the above-mentioned diseases. Here I introduce primary and iPSC-derived organoids as well as computational models to predict different metabolic outcomes. Organoids The development discovery of organoids in 200998 has entailed a change in the way we conceive the study of biology and biomedicine. Organoids are three-dimensional structures that can be derived from primary tissue as well
18 Chapter 1 as from adult or pluripotent stem cells. They have the ability to proliferate in vitro, while maintaining some of the functionalities of the organ of origin99. These structures have provided scientists with a revolutionary tool which allowed in vitro research to get one step closer to physiological relevance. While traditional 2D cell cultures rely on monolayers of one cell type, organoids allow more than one cell type to communicate with each other in a way that recapitulates the structure and functions of the tissue of origin more closely100. Another advantage of organoids over 2D cultures is the possibility of deriving patient-specific organoids, allowing for more tailored research101. Ever since their discovery, many organoid systems have been described, recapitulating the functional and structural properties of multiple organs, including intestine98, liver102,103, stomach104 and brain105. Organoids have been used for many purposes, including understanding organ development, modeling of genetic diseases, drug screening, regenerative medicine and the study of organ-specific metabolism99,106–108. While organoids are an excellent tool to study the effect of genetic diseases, as illustrated by an extensive number of publications109, they are also highly interesting for the study of different nutritional stressors on organ metabolism110. Moreover, they can also be used to understand the effects of different dietary interventions on the organ of study and its metabolic regulation111. As mentioned above, organoids can be isolated either from primary tissue or derived from stem cells, both adult and pluripotent99. Development of induced pluripotent stem cells (iPSCs) in 2006 by Yamanaka was regarded as a breakthrough in the field of stem cell biology112. Ever since, numerous protocols describing the differentiation of iPSCs into multiple cellular lineages have been published113–116. The hepatic lineage has been induced from iPSCs using multiple protocols which are commonly based on the use of growth factors Activin A and BMP4 to induce the definitive endoderm, followed by a combination of other growth factors to further commit the cells into hepatoblasts113,114,117–119. iPSCderived hepatocytes and hepatobiliary organoids recapitulate many features of the liver such as albumin production, lipid accumulation, and urea production. In some cases, upon further differentiation towards the hepatic-lineage, these organoids are referred to as “hepatocyte-like”. iPSC-derived hepatobiliary and hepatocyte-like organoids are, however, limited in their maturity level and are closer to fetal than to adult liver120,121. iPSC-derived hepatobiliary organoid models have been used to study different diseases such as alcoholic liver injury119 or genetic diseases such as Alagille syndrome (ALGS) and Tetralogy of Fallot (TOF)122.
19 1 General Introduction Systems medicine approach and computational models Systems medicine is the multidisciplinary combination of different fields including biology, informatics, computational modeling and mathematics to gap the bridge between new translational research and traditional health care123. This interdisciplinary methodology allows clinical investigators and physicians to team up with mathematicians and natural scientists, to tackle different medical problems in a more efficient and tailored manner124. Their approaches commonly rely on in vitro and in vivo systems, combined with various ‘omics technologies as well as different computational approaches (in silico systems). Computational models are in silico representations of biological systems that can help us understand regulation of different biological processes. These models mathematically describe a biological system in an integrative way. Simulating their response to interventions or mutations can help us predict behaviors and outcomes of a given system under different conditions. In this thesis I will focus on computational models of metabolic pathways. One main type of metabolic computational models is a detailed kinetic model. Kinetic models for the study of metabolic pathways are based on kinetic equations describing the behavior of the enzymes of a given pathway combined with kinetic parameters commonly obtained from the literature or measured in the lab. Ordinary differential equations (ODEs) are then used to describe the dynamics of reactions in the system and how these produce and consume different metabolites. These dynamic models can predict both metabolite concentrations and fluxes over time. An interesting aspect of detailed kinetic models is the possibility to study Metabolic Control Analysis to understand the control that certain enzymes exert over metabolite concentrations or metabolic fluxes in a specific pathway125,126. In order to adapt computational models to different conditions, such as different tissues or diseases, enzyme concentrations can be measured and introduced into the model. To do so, proteomics data or enzyme activity data are commonly used. There are models describing different metabolic pathways such as mitochondrial β-oxidation95, TCA and ECT127 and glucose metabolism128. So far, to the best of my knowledge, no computational models describing peroxisomal β-oxidation have been published. Another highly valuable type of computational model for the study of systems medicine is based on deep learning (DL) algorithms. These models are often referred to as ‘multilayered’ due to their ability to process data in different layers in which they transform the data and pass them on to the next layer until
20 Chapter 1 the output layer129. They rely on big data sets and can be used to automatize analyses and quantification of different parameters. In the context of this thesis, I will present an application of DL for the analysis of organoid number and size from brightfield microscopy images. AIMS AND OUTLINE OF THIS THESIS The aim of this thesis is to develop and characterize in vitro and computational translational models to study two diseases affecting fatty acid metabolism. In both diseases, peroxisomal and mitochondrial homeostasis are known or hypothesized to be disrupted. In the case of malnutrition, it has been proposed that peroxisomal loss precedes the mitochondrial phenotype56. In the case of MCADD, peroxisomes have been hypothesized to play a compensatory role in the disease. Experimental and computational tools to study these two diseases largely overlap. Studying two different diseases, in which the interplay between mitochondria and peroxisomes is affected, will provide deeper insight into the underlying mechanisms. In part I, I focus on the development and characterization of two in vitro malnutrition models from mouse primary tissue (both liver and intestine), based on growth media lacking amino acids. In part II, I first make use of this newly characterized malnutrition hepatic organoid model to understand the underlying causes for peroxisomal loss under low amino acid conditions. Secondly, I test different pharmacological approaches to prevent peroxisomal loss in the liver. Moreover, I also develop a detailed kinetic model of the peroxisomal β-oxidation to understand the effects of malnutrition and the different treatments on the pathway. In part III, I focus on the study of MCADD. To do so, I develop and characterize an MCADD in vitro model making use of patient specific iPSC-derived hepatic organoids and asses the role peroxisomes play in the disease. Finally, in part IV I focus on contextualizing the results presented throughout the thesis and discuss the literature on the topic. I also discuss the limitations and possibilities of these new translational models (both from the technical and the biological point of view) and highlight and potential ways to improve in the future. Part I – Malnutrition in vitro In chapter 2, I develop and characterize an organoid in vitro model to study malnutrition both in the liver and the intestine. These 3D systems replicate many of the organ-specific malnutrition phenotypes of both organs. In the case
21 1 General Introduction of the liver, amino-acid deprivation leads to hypoalbuminemia and hepatic steatosis, while the effect on the intestine is characterized by increased barrier permeability. Both organs show a clear reduction in peroxisomal protein markers and dysfunctional mitochondria which are thought to be responsible for the hepatic steatosis and disruption of the intestinal barrier permeability. Moreover, as a proof of principle, I demonstrate how these in vitro models can be used as a pre-screening tool to test potential treatments to prevent both peroxisomal and mitochondrial loss. Fenofibrate and rapamycin are used to prevent peroxisomal loss and mitochondrial loss in the liver and the intestine respectively. In chapter 3 I make use of the opportunities offered by the embedding of my PhD project in the European Perico training network, aimed at deciphering the role of peroxisomes in cellular interaction and signaling. In this chapter one of my Perico colleagues develops and implements a deep learning computational model to measure and track organoid size and growth. Together, we apply this tool to detect and measure different types of cystic organoids in brightfield microscopy images. It can be used for a wide range of purposes in different aspects of organoid research. In order to illustrate the potential of this tool we use it to measure the size of organoids under control and amino-acid deprived conditions, and we compare the results with manually annotated data from chapter 2. Part II- Applications of the new malnutrition models In chapter 4, I study the mechanisms underlying peroxisomal loss and potential treatments to prevent such loss. To do so, I make use of the hepatic in vitro malnutrition model described in chapter 2. Together with a colleague from the pediatrics department at the UMCG I demonstrate that amino-acid deprivation in hepatic organoids is accompanied by a clear induction of the autophagic flux, which we measure using a novel autophagic probe. Furthermore, I apply this model to identify different therapeutic compounds. Finally, I show that longchain poly-unsaturated fatty acids (LCPUFA) are more effective than synthetic PPAR-α agonists and demonstrate that docosahexaenoic acid (DHA) prevents peroxisomal and mitochondrial loss during amino-acid deficiency. In chapter 5, I develop a kinetic model of the peroxisomal β-oxidation based on kinetic studies described in the literature. Doing so, I review the kinetics of the enzymes involved in the pathway. Then I use the model to predict the effects of amino acid deficiency in the pathway. For that purpose, I combine targeted proteomics data from amino-acid deprived organoids and the above-
22 Chapter 1 mentioned kinetic model. Moreover, I also analyze the effects of the previously described DHA supplement on the pathway to understand the role of this PPAR-α agonist. Part III- Medium Chain Acyl-CoA Dehydrogenase Deficiency Next, in chapter 6 together with my colleague from the Systems Medicine group of the UMCG I develop an iPSC-derived hepatobiliary organoid model to study a particular IEM. In this case, we use fibroblasts-derived iPSCs from healthy and MCAD-deficient children and differentiate them into hepatocyte-like organoids. We show that these organoids recapitulate the biochemical MCADD phenotype and serve as valuable tool to understand pathophysiological aspects of the disease. Finally, once the MCADD organoids have been established and characterized we focus on the study of peroxisomes and the potential role they play in metabolizing medium-chain fatty acids (accumulated in the absence of MCAD) as well as on their role in coenzyme A (CoA) metabolism. Part IV- General Discussion and future perspectives Finally, in chapter 7 I evaluate the progress made in the study of malnutrition and MCAD deficiency with this thesis and the advances it permitted. I focus on the potential and limitations of in vitro models and propose different solutions. I also focus on the interplay of peroxisomes and mitochondria in health and disease and the potential compensatory mechanisms from one organelle to the other. Finally, I discuss the future directions and application of translational models for metabolic research.
23 1 General Introduction REFERENCES 1. De Carvalho, C. C. C. R. & Caramujo, M. J. The various roles of fatty acids. Molecules vol. 23 Preprint at https://doi.org/10.3390/molecules23102583 (2018). 2. Houten, S. M., Violante, S., Ventura, F. V. & Wanders, R. J. A. The Biochemistry and Physiology of Mitochondrial Fatty Acid β-Oxidation and Its Genetic Disorders. Annual Review of Physiology vol. 78 23–44 Preprint at https://doi.org/10.1146/annurevphysiol-021115-105045 (2016). 3. Spinelli, J. B. & Haigis, M. C. The multifaceted contributions of mitochondria to cellular metabolism. Nature Cell Biology vol. 20 745–754 Preprint at https://doi.org/10.1038/ s41556-018-0124-1 (2018). 4. Prosad Chakrabarty, R. & Chandel, N. S. Beyond ATP, new roles of mitochondria. Biochem (London) 44, 2–8 (2022). 5. Chan, D. C. Mitochondrial Dynamics and Its Involvement in Disease. (2019) doi:10.1146/ annurev-pathmechdis. 6. Fenton, A. R., Jongens, T. A. & Holzbaur, E. L. F. Mitochondrial dynamics: Shaping and remodeling an organelle network. Current Opinion in Cell Biology vol. 68 28–36 Preprint at https://doi.org/10.1016/j.ceb.2020.08.014 (2021). 7. Eisner, V., Picard, M. & Hajnóczky, G. Mitochondrial dynamics in adaptive and maladaptive cellular stress responses. Nature Cell Biology vol. 20 755–765 Preprint at https://doi.org/10.1038/s41556-018-0133-0 (2018). 8. Wanders, R. J. A., Baes, M., Ribeiro, D., Ferdinandusse, S. & Waterham, H. R. The physiological functions of human peroxisomes. Physiological Reviews vol. 103 957–1024 Preprint at https://doi.org/10.1152/physrev.00051.2021 (2023). 9. Michels, P. A. M. & Gualdrón-López, M. Biogenesis and metabolic homeostasis of trypanosomatid glycosomes: New insights and new questions. Journal of Eukaryotic Microbiology vol. 69 Preprint at https://doi.org/10.1111/jeu.12897 (2022). 10. Weller, S., Gould, S. J. & Valle, D. Peroxisome Biogenesis Disorders. Annual Review of Genomics and Human Genetics vol. 4 165–211 Preprint at https://doi.org/10.1146/ annurev.genom.4.070802.110424 (2003). 11. Waterham, H. R., Ferdinandusse, S. & Wanders, R. J. A. Human disorders of peroxisome metabolism and biogenesis. Biochim Biophys Acta Mol Cell Res 1863, 922–933 (2016). 12. Schrader, M., Costello, J. L., Godinho, L. F., Azadi, A. S. & Islinger, M. Proliferation and fission of peroxisomes - An update. Biochimica et Biophysica Acta - Molecular Cell Research vol. 1863 971–983 Preprint at https://doi.org/10.1016/j.bbamcr.2015.09.024 (2016). 13. Schrader, M., Kamoshita, M. & Islinger, M. Organelle interplay—peroxisome interactions in health and disease. Journal of Inherited Metabolic Disease vol. 43 71–89 Preprint at https://doi.org/10.1002/jimd.12083 (2020). 14. Schrader, M., Bonekamp, N. A. & Islinger, M. Fission and proliferation of peroxisomes. Biochimica et Biophysica Acta - Molecular Basis of Disease vol. 1822 1343–1357 Preprint at https://doi.org/10.1016/j.bbadis.2011.12.014 (2012).
24 Chapter 1 15. Wróblewska, J. P. & van der Klei, I. J. Peroxisome maintenance depends on de novo peroxisome formation in yeast mutants defective in peroxisome fission and inheritance. Int J Mol Sci 20, (2019). 16. Banerjee, S. & Prinz, W. A. Early steps in the birth of four membrane-bound organelles—Peroxisomes, lipid droplets, lipoproteins, and autophagosomes. Curr Opin Cell Biol 84, (2023). 17. Violante, S. et al. Peroxisomes can oxidize medium- and long-chain fatty acids through a pathway involving ABCD3 and HSD17B4. FASEB Journal 33, 4355–4364 (2019). 18. Osmundsen, H., Neat, C. E. & Borrebaek, B. Fatty acid products of peroxisomal β-oxidation. J Bmhum vol. 12. 19. Schrader, M. & Yoon, Y. Mitochondria and peroxisomes: Are the ‘Big Brother’ and the ‘Little Sister’ closer than assumed? BioEssays vol. 29 1105–1114 Preprint at https://doi. org/10.1002/bies.20659 (2007). 20. Fransen, M., Lismont, C. & Walton, P. The peroxisome-mitochondria connection: How and why? International Journal of Molecular Sciences vol. 18 Preprint at https://doi. org/10.3390/ijms18061126 (2017). 21. Wanders, R. J. A., Waterham, H. R. & Ferdinandusse, S. Metabolic interplay between peroxisomes and other subcellular organelles including mitochondria and the endoplasmic reticulum. Frontiers in Cell and Developmental Biology vol. 3 Preprint at https://doi.org/10.3389/fcell.2015.00083 (2016). 22. Shields, H. J., Traa, A. & Van Raamsdonk, J. M. Beneficial and Detrimental Effects of Reactive Oxygen Species on Lifespan: A Comprehensive Review of Comparative and Experimental Studies. Front Cell Dev Biol 9, (2021). 23. Fransen, M., Nordgren, M., Wang, B. & Apanasets, O. Role of peroxisomes in ROS/ RNS-metabolism: Implications for human disease. Biochimica et Biophysica Acta - Molecular Basis of Disease vol. 1822 1363–1373 Preprint at https://doi.org/10.1016/j. bbadis.2011.12.001 (2012). 24. Neuspiel, M. et al. Cargo-Selected Transport from the Mitochondria to Peroxisomes Is Mediated by Vesicular Carriers. Current Biology 18, 102–108 (2008). 25. Horner, S. M., Liu, H. M., Park, H. S., Briley, J. & Gale, M. Mitochondrial-associated endoplasmic reticulum membranes (MAM) form innate immune synapses and are targeted by hepatitis C virus. Proc Natl Acad Sci U S A 108, 14590–14595 (2011). 26. Antonenkov, V. D. & Hiltunen, J. K. Transfer of metabolites across the peroxisomal membrane. Biochimica et Biophysica Acta - Molecular Basis of Disease vol. 1822 1374– 1386 Preprint at https://doi.org/10.1016/j.bbadis.2011.12.011 (2012). 27. Camões, F., Bonekamp, N. A., Delille, H. K. & Schrader, M. Organelle dynamics and dysfunction: A closer link between peroxisomes and mitochondria. in Journal of Inherited Metabolic Disease vol. 32 163–180 (2009). 28. Fourcade, S., Ferrer, I. & Pujol, A. Oxidative stress, mitochondrial and proteostasis malfunction in adrenoleukodystrophy: A paradigm for axonal degeneration. Free Radical Biology and Medicine vol. 88 18–29 Preprint at https://doi.org/10.1016/j. freeradbiomed.2015.05.041 (2015). 29. Salpietro, V. et al. Zellweger syndrome and secondary mitochondrial myopathy. Eur J Pediatr 174, 557–563 (2015).
25 1 General Introduction 30. Góth, L. & Nagy, T. Inherited catalase deficiency: Is it benign or a factor in various age related disorders? Mutation Research - Reviews in Mutation Research vol. 753 147–154 Preprint at https://doi.org/10.1016/j.mrrev.2013.08.002 (2013). 31. Peeters, A. et al. Mitochondria in peroxisome-deficient hepatocytes exhibit impaired respiration, depleted DNA, and PGC-1α independent proliferation. Biochim Biophys Acta Mol Cell Res 1853, 285–298 (2015). 32. Mardinoglu, A., Uhlen, M. & Borén, J. Broad Views of Non-alcoholic Fatty Liver Disease. Cell Systems vol. 6 7–9 Preprint at https://doi.org/10.1016/j.cels.2018.01.004 (2018). 33. Kneeman, J. M., Misdraji, J. & Corey, K. E. Secondary causes of nonalcoholic fatty liver disease. Therapeutic Advances in Gastroenterology vol. 5 199–207 Preprint at https:// doi.org/10.1177/1756283X11430859 (2012). 34. Lopez-Pedrosa, J. M., Torres, M. I., Ferná ndez, M. I., RıB os, A. & Gil, A. Nutrient Metabolism Severe Malnutrition Alters Lipid Composition and Fatty Acid Profile of Small Intestine in Newborn Piglets 1,2. J. Nutr vol. 128 (1998). 35. Pezeshki, A. & Chelikani, P. K. Low Protein Diets and Energy Balance: Mechanisms of Action on Energy Intake and Expenditure. Frontiers in Nutrition vol. 8 Preprint at https://doi.org/10.3389/fnut.2021.655833 (2021). 36. Thompson, D. S. et al. The effect of wasting and stunting during severe acute malnutrition in infancy on insulin sensitivity and insulin clearance in adult life. J Dev Orig Health Dis 13, 750–756 (2022). 37. Francis-Emmanuel, P. M. et al. Glucose metabolism in adult survivors of severe acute malnutrition. Journal of Clinical Endocrinology and Metabolism 99, 2233–2240 (2014). 38. Khardori, R., Bajaj, J. S., Deo, M. G. & Bansal, O. D. Insulin secretion and carbohydrate metabolism in experimental protein malnutrition. J. Endocrinol. Invest vol. 3 (1980). 39. Kitada, M., Ogura, Y., Monno, I. & Koya, D. The impact of dietary protein intake on longevity and metabolic health. EBioMedicine vol. 43 632–640 Preprint at https://doi. org/10.1016/j.ebiom.2019.04.005 (2019). 40. Solon-Biet, S. M. et al. The ratio of macronutrients, not caloric intake, dictates cardiometabolic health, aging, and longevity in ad libitum-fed mice. Cell Metab 19, 418–430 (2014). 41. Garg, U. & Smith, L. D. Introduction to laboratory diagnosis and biomarkers in inborn error of metabolism. in Biomarkers in Inborn Errors of Metabolism 1–24 (Elsevier, 2017). doi:10.1016/b978-0-12-802896-4.00001-8. 42. Wanders, R. J. A., Visser, G., Ferdinandusse, S., Vaz, F. M. & Houtkooper, R. H. Mitochondrial fatty acid oxidation disorders: Laboratory diagnosis, pathogenesis, and the complicated route to treatment. Journal of Lipid and Atherosclerosis vol. 9 313–333 Preprint at https://doi.org/10.12997/jla.2020.9.3.313 (2020). 43. World Health Organization. What is malnutrition? [Internet]. 2021. Available from: https://www.who.int/news-room/fact-sheets/detail/malnutrition. 44. Von Grebmer, K. et al. Synopsis: 2014 global hunger index: The challenge of hidden hunger. vol. 83 ( Intl Food Policy Res Inst, 2014).
26 Chapter 1 45. WHO, World Bank. United Nations Children’s Fund, World Health Organization & World Bank Group. Levels and trends in child malnutrition: Key findings of the 2023 Edition of the Joint Child Malnutrition Estimates [Internet]. Geneva: WHO. 2023 Available from: https://www.who.int/publications/i/item/9789240073791. 46. Briend, A. Collaborating to improve the management of acute malnutrition worldwide Kwashiorkor: still an enigma-the search must go on. www.cmamforum.org (2014). 47. Bhutta, Z. A. et al. Severe childhood malnutrition. Nat Rev Dis Primers 3, (2017). 48. Grenov, B. et al. Diarrhea, Dehydration, and the Associated Mortality in Children with Complicated Severe Acute Malnutrition: A Prospective Cohort Study in Uganda. Journal of Pediatrics 210, 26-33.e3 (2019). 49. Attia, S., Feenstra, M., Swain, N., Cuesta, M. & Bandsma, R. H. J. Starved Guts: Morphologic and Functional Intestinal Changes in Malnutrition. Journal of Pediatric Gastroenterology and Nutrition vol. 65 491–495 Preprint at https://doi.org/10.1097/ MPG.0000000000001629 (2017). 50. Bandsma, R. H. J. et al. Mechanisms Behind Decreased Endogenous Glucose Production in Malnourished Children. Pediatr Res. (2010) doi:10.1203/PDR.0b013e3181f2b959. 51. Wharton, B. Hypoglycaemia in children with Kwashiorkor. The Lancet 124, 617 (1970). 52. Badaloo, A. V, Forrester, T., Reid, M. & Jahoor, F. Lipid kinetic differences between children with kwashiorkor and those with marasmus. Am J Clin Nutr. (2006) doi:10.1093/ ajcn/83.6.1283. 53. Badaloo, A., Reid, M., Soares, D., Forrester, T. & Jahoor, F. Relation between liver fat content and the rate of VLDL apolipoprotein B-100 synthesis in children with proteinenergy malnutrition. Am J Clin Nutri. (2005) doi:10.1093/ajcn/81.5.1126. 54. Doherty, J. F., Golden, M. H. & Brooks, S. E. Peroxisomes and the fatty liver of malnutrition: an hypothesis. Am J Clin Nuir 54, 674–681 (1991). 55. Enwonwu, C. 0, Worthington, B. S. & Jacobson, K. L. Protein-Energy Malnutrition in Infant Non-Human Primates (Macaca Nemestrina) I. Correlation of Biochemical Changes with Fine Structural Alterations in the Liver. Br. J. exp. Path 58, 78 (1977). 56. van Zutphen, T. et al. Malnutrition-associated liver steatosis and ATP depletion is caused by peroxisomal and mitochondrial dysfunction. J Hepatol 65, 1198–1208 (2016). 57. Arvidsson Kvissberg, M. E. et al. Inhibition of mTOR improves malnutrition induced hepatic metabolic dysfunction. Sci Rep 12, (2022). 58. Hu, G. et al. The role of the tryptophan-NAD + pathway in a mouse model of severe malnutrition induced liver dysfunction. Nat Commun 13, (2022). 59. Mast, F. D., Rachubinski, R. A. & Aitchison, J. D. Signaling dynamics and peroxisomes. Current Opinion in Cell Biology vol. 35 131–136 Preprint at https://doi.org/10.1016/j. ceb.2015.05.002 (2015). 60. Ling, C. et al. Rebalancing of mitochondrial homeostasis through an NAD +-SIRT1 pathway preserves intestinal barrier function in severe malnutrition. EBioMedicine 96, (2023). 61. Glick, D., Barth, S. & Macleod, K. F. Autophagy: Cellular and molecular mechanisms. Journal of Pathology vol. 221 3–12 Preprint at https://doi.org/10.1002/path.2697 (2010). 62. Mizushima, N. SnapShot: Organelle degradation. Molecular Cell vol. 82 1604-1604.e1 Preprint at https://doi.org/10.1016/j.molcel.2022.03.015 (2022).
www.ridderprint.nlRkJQdWJsaXNoZXIy MTk4NDMw