The Pennsylvania State University
The Graduate School
College of Medicine
DOES THE H67D HFE GENOTYPE IMPACT MACROPHAGE
PHENOTYPE AND DISEASE PROCESS?
A Dissertation in
Anatomy
by
Anne Marie Nixon
Ó 2017 Anne Marie Nixon
Submitted in Partial Fulfillment
of the Requirements
for the Degree of
Doctor of Philosophy
December 2017
The dissertation of Anne Marie Nixon was reviewed and approved* by the following:
James R. Connor University Distinguished Professor Vice-Chair of the Department of Neurosurgery Dissertation Advisor Chair of Committee
Gregory M. Holmes Associate Professor of Neural and Behavioral Sciences
Robert F. Paulson Professor of Veterinary and Biomedical Sciences
Patricia J. McLaughlin Professor of Neural and Behavioral Sciences Director of the Graduate Program in Anatomy
*Signatures are on file in the Graduate School
iii ABSTRACT
Iron homeostasis is tightly regulated to prevent disease resulting from too little or too much iron within the body. The HFE protein is involved in iron regulation and when mutated it is associated with several iron-related disorders. In particular, the H63D HFE mutant variant is the most prevalent HFE mutation in the Caucasian population. Although HFE is known to modulate iron homeostasis through its interaction with transferrin receptor (TfR), its precise role in iron metabolism and cellular function remain unclear. One cell type in the body whose function may be especially impacted by the HFE mutation is the macrophage. Macrophages play critical roles in iron regulation through scavenging and recycling of iron, providing a mechanism to both reduce and reallocate excess iron. The effects of perturbed iron regulation in macrophages in neurodegenerative disease progression is not yet established. Here, this thesis aims to investigate the role of HFE in macrophage function, and determine how HFE mutant macrophages contribute to neurodegenerative diseases, specifically Parkinson’s Disease. We hypothesize that the H67D macrophages would have altered iron handling and inflammatory response, promoting
Parkinson’s progression.
To investigate the role of H67D genotype on macrophage phenotype, bone marrow macrophages from H67D mutant mice were generated and compared to wildtype derived macrophages for the status of iron related proteins, inflammatory cytokine secretion, and macrophage-related cell functions (i.e. migration and phagocytosis). The H67D macrophages had increased levels of L-ferritin, but no significant differences in TfR or ferroportin, indicating differences in iron handling specifically in the resting state. In addition, H67D macrophages functionality was altered as they displayed decreased migration, and increased phagocytosis
iv activity compared to WT. The H67D macrophages also expressed increased levels of chemoattractant proteins and decreased inflammatory cytokines.
The alterations in macrophage function with the HFE mutation could impact the role of these cells in neurodegenerative disease. Therefore, we generated a Parkinsonism mouse model using paraquat to selectively target and induce the degeneration of dopaminergic neurons within the nigrostriatal pathway. In this model, we found WT mice had impaired motor function, and decreased number of tyrosine hydroxylase (TH) neurons, whereas the H67D mice showed no motor impairment following exposure to paraquat. Surprisingly, the mice with the H67D mutation had a significant reduction in TH neurons in both treated and untreated mice compared to WT. Thus, the HFE mutation was associated with significant loss of TH neurons but this loss was not increased by paraquat. The loss of these neurons could be the result of a developmental flaw or due to an initial stressor that led to neuroprotection; which is discussed in the final chapter.
The third chapter used the same paraquat model to determine if there was a genotype effect of cholesterol or statin exposure. Cholesterol and statin have been proposed as agents that progress Parkinson’s Disease. Moreover, the H63D/H67D mutation is associated with disruption in cholesterol metabolism. The results showed that the cholesterol diet negatively affected motor function of WT mice when injected with paraquat, however there was no effect on motor function with a statin diet. The H67D mice have no change in motor function when on either a cholesterol or statin diet indicating the H67D genotype is resistant to the toxic effects of cholesterol. In addition, a statin diet does not exacerbate the results of paraquat induced neurotoxicity.
v Our results show that the H67D genotype plays a role in iron and immune response of macrophages but also can be neuroprotective in a Parkinsonism disease model possibly due to the initial stress exerted by the macrophages. Overall, these findings encompass a complete role of
H67D from periphery to a disease model. The H67D mouse model can further be used to interrogate the mechanism for the neuroprotection against exposure to paraquat and to cholesterol plus paraquat. These data further support the concept that the HFF genotype is a disease modifier.
vi TABLE OF CONTENTS
LIST OF FIGURES ...... ix
LIST OF TABLES ...... xi
LIST OF ABBREVIATIONS ...... xii
ACKNOWLEDGEMENTS ...... xiv
Chapter 1 Does HFE Genotype Impact Macrophage Phenotype in Disease Process and Therapeutic Response? ...... 1
1.1 Abstract ...... 1 1.2 Iron ...... 1 1.2.1 Iron Regulation ...... 2 1.3 Hemochromatosis ...... 3 1.3.1 Discovery ...... 3 1.3.2 Types ...... 4 1.4 HFE ...... 5 1.4.1 Structure ...... 5 1.4.2 Function ...... 5 1.4.3 Polymorphisms ...... 6 1.4.4 HFE and Neurodegenerative Diseases ...... 8 1.4.4.1 Alzheimer's Disease ...... 9 1.4.4.2 Amyotrophic Lateral Sclerosis ...... 9 1.4.4.3 Parkinson's Disease ...... 10 1.4.5 HFE and Cancer ...... 11 1.4.6 HFE Animal Models ...... 12 1.5 Macrophages ...... 13 1.5.1 Macrophages and Iron Metabolism ...... 14 1.5.2 Macrophage Polarization ...... 14 1.5.3 Macrophages and HFE ...... 16 1.6 Conclusions ...... 17 1.7 Acknowledgements ...... 17 1.8 References ...... 18
Chapter 2 Research Gaps ...... 40
2.1 Research Gaps ...... 40 2.2 References ...... 46
Chapter 3 The Role of HFE Genotype in Macrophage Phenotype ...... 49
3.1 Abstract ...... 49 3.2 Introduction ...... 50 3.3 Materials and Methods ...... 51
vii 3.3.1 Mouse Colony ...... 51 3.3.2 Primary Macrophage Culture ...... 52 3.3.3 MTT and LDH Assays ...... 52 3.3.4 Enzyme Linked Immunosorbent Assays (ELISAs) ...... 53 3.3.4.1 Iron Proteins ...... 53 3.3.4.1 Bone Morphogenetic Protein 6 (BMP6) and Bone Morphogenetic Type I Receptor (ALK3) ...... 53 3.3.5 Phagocytosis Assay ...... 54 3.3.6 Comparison of Cellular Migration ...... 54 3.3.7 Analysis of Macrophage Cytokines ...... 55 3.3.8 59Fe Loading and Release ...... 55 3.3.9 Statistical Analysis ...... 56 3.4 Results ...... 56 3.4.1 Cellular Proliferation of Wildtype Macrophages Increases in Iron Rich Conditions ...... 56 3.4.2 Iron Loading Promotes L-ferritin Expression in H67D Mutant BMMs ...... 58 3.4.3 H67D Mutant Promotes BMP6 Secretion ...... 61 3.4.4 Radioactive Iron Release ...... 63 3.4.5 H67D Regulates Migration, Phagocytosis and Inflammatory Cytokine Expression in BMMs ...... 65 3.5 Discussion ...... 70 3.6 References ...... 75
Chapter 4 The Neuroprotective Role of the HFE Gene Variant H67D in a Paraquat Mouse Model ...... 81
4.1 Abstract ...... 81 4.2 Introduction ...... 82 4.3 Materials and Methods ...... 84 4.3.1 Animals ...... 84 4.3.2 Genotyping ...... 84 4.3.3 Paraquat and Saline Injections ...... 85 4.3.4 Behavior ...... 87 4.3.5 Histology ...... 87 4.3.5.1 Immunohistochemistry ...... 87 4.3.5.2 Iron Staining ...... 88 4.3.6 Magnetic Resonance Imaging ...... 88 4.3.6.1 Image Acquisition ...... 88 4.3.6.2 MRI Parametric Analysis ...... 89 4.3.7 Statisitical Analysis ...... 89 4.4 Results ...... 90 4.4.1 Effect of Paraquat on Motor Function ...... 92 4.4.2 Tyrosine Hydroxylase Neuron Loss ...... 95 4.4.3 L-ferritin Levels within the Substantia Nigra ...... 98 4.4.4 Microglia and Astrocyte Changes within Substantia Nigra ...... 101 4.4.5 Iron Accumulation within the Substantia Nigra ...... 104 4.5 Discussion ...... 106 4.6 Acknowledgements ...... 109
viii 4.7 References ...... 110
Chapter 5 The Effect of Cholesterol and Statin Diet on Motor Function in H67D Paraquat Mouse Model ...... 117
5.1 Abstract ...... 117 5.2 Introduction ...... 117 5.3 Materials and Methods ...... 120 5.3.1 Animals ...... 120 5.3.2 Experimental Conditions ...... 120 5.3.3 Evaluation of Motor Function ...... 121 5.3.4 Statistical Analysis ...... 121 5.4 Results ...... 121 5.4.1 Comparison of Diet Composition ...... 124 5.4.2 Changes in Body Weight and Mean Food Intake ...... 126 5.4.3 Behavioral Differences between Genotype, Treatment, and Diet Groups ...... 128 5.4.4 Behavioral Differences Between Diet and Treatment, and Treatment and Genotype ...... 130 5.5 Discussion ...... 133 5.6 Acknowledgements ...... 134 5.7 References ...... 135
Chapter 6 Underlying Mechanisms and Translational Implications of the H67D Mutation; Future Directions ...... 140
6.1 Summary of Main Finding ...... 140 6.2 A Little Stress Exerted by H67D can be Neuroprotective ...... 142 6.3 The Varying Role of H67D in an ALS Mouse Model ...... 148 6.4 Macrophage Therapy in Parkinson's Disease ...... 150 6.5 The Potential Role of C282Y in Macrophage Phenotype and Disease Progression .. 151 6.6 Does the H67D Genotype Exert Long Term Neuroprotection in Paraquat Mouse Model? ...... 152 6.7 The Role of Cholesterol and Statins in H67D Macrophages ...... 153 6.8 Bone Marrow Macrophages vs. Peritoneal Macrophages ...... 155 6.9 Circuitry of the Substantia Nigra ...... 156 6.10 Conclusions ...... 156 6.11 References ...... 158
Appendix Peritoneal Macrophage L-ferritin Concentration ...... 165
ix LIST OF FIGURES
Figure 2-1: Hypothesized Chapter 3 Results...... 41
Figure 2-2: Hypothesized Chapter 4 Results...... 43
Figure 2-3: Hypothesized Chapter 5 Results...... 44
Figure 3-1: Cell Proliferation (MTT) and Cytotoxicity (LDH) of BMMs Incubated with Control, FAC, or DFO Media...... 57
Figure 3-2: Iron Handling Protein Concentrations in Cellular Lysates...... 59
Figure 3-3: BMP6 and BMP6 Receptor, ALK3, Concentration in Cellular Culture Media and Cellular Lysates...... 62
Figure 3-4: Amount of 59Fe released from BMMs following 24 hours of 59Fe loading...... 64
Figure 3-5: Cellular Migration of the Chemoattractant FBS...... 66
Figure 3-6: Resting State Phagocytosis Activity...... 67
Figure 3-7: Pro-inflammatory Cytokine Array Panel...... 69
Figure 4-1: Schematic Timeline of the Study Design...... 86
Figure 4-2: Mean Food Intake and Change in Body Weight...... 91
Figure 4-3: Behavior Analysis of Motor Function between Genotype and Treatment Groups...... 93
Figure 4-4: MRI Analysis of Genotype and Treatment Groups...... 94
Figure 4-5: Tyrosine Hydroxylase (TH) Expression in the Substantia Nigra...... 96
Figure 4-6: L-ferritin Expression in the Substantia Nigra...... 99
Figure 4-7: Microglia Expression in the Substantia Nigra...... 102
Figure 4-8: Astrocyte Expression in the Substantia Nigra...... 103
Figure 4-9: Iron (Perl) Expression in the Substantia Nigra...... 105
Figure 5-1: Schematic Timeline of Experimental Design...... 123
Figure 5-2: Comparison of Diet Composition...... 125
Figure 5-3: Experimental Group Comparison of Body Weight and Food Intake ...... 127
x Figure 5-4: Motor Behavioral Differences between Genotype, Treatment, and Diet Groups ...... 129
Figure 5-5: Effect of Diet and Treatment on Behavioral Motor Function ...... 131
Figure 5-6: Effect of Treatment and Genotype on Behavioral Motor Function...... 132
Figure 7-1: L-ferritin Concentration within Peritoneal Macrophage Cellular Lysates...... 165
xi
LIST OF TABLES
Table 6-1: Experimental Design of H67D Macrophage Conditioned Media on Dopaminergic Neurons...... 147
xii
LIST OF ABBREVIATIONS
Ab amyloid-b
AD Alzheimer’s Disease
ALK3/BMPI Bone Morphogenetic Protein Type I Receptor
ALL Acute Lymphocytic Leukemia
ALS Amyotrophic Lateral Sclerosis
ANOVA Analysis of Variance
Apo-Tf Apo-Transferrin (Iron Poor Transferrin)
APP Amyloid Precursor Protein b2M b2 Microglobulin
BMM Bone Marrow Macrophage
BMP Bone Morphogenetic Protein
DFO Deferoxamine
DPBS Dulbecco’s Phosphate-Buffered Saline
ELISA Enzyme-Linked Immunosorbent Assay
FAC Ferric Ammonium Citrate
FBS Fetal Bovine Serum
Fe2+ Ferrous Iron
Fe3+ Ferric Iron
HJV Hemojuvelin
HO Heme Oxygenase
IFNg Interferon-g
xiii IMP Iron Management Protein
IRE Iron Regulatory Element
IRP Iron Regulatory Protein
LDH Lactate Dehydrogenase
LPS Lipopolysaccharide
M-CSF Macrophage-Colony Stimulating Factor
MCP-1 Monocyte Chemoattractant Protein-1
MRI Magnetic Resonance Imaging
MTT 3-(4,5-Dimethylthiazol-2-Yl)-2,5-Diphenyltetrazolium Bromide
PD Parkinson’s Disease
ROS Reactive Oxygen Species
SD Standard Deviation
SEM Standard Error of Mean
STEAP-3 Six Transmembrane Epithelial Antigen of the Prostate-3
Tf Transferrin
TAM Tumor Associated Macrophage
TH Tyrosine Hydroxylase
UTR Untranslated Region
WT Wildtype
xiv
ACKNOWLEDGEMENTS
I dedicate this thesis to my husband Chris, who has supported me unconditionally throughout this journey. Your undying attention, love, and encouragement has made the good times better, and hard times the easier. Your patience is a virtue and I will forever be grateful for everything you have done for me.
I want to especially thank my family and friends, who have provided me with support, encouragement, and inspiration. To my mom this thesis is dedicated to you as well. You have always believed in me and supported all of my dreams. I would not have been able to get through this without you and all of your phone calls. To my dad and brother, Phillip, thank you for all of your love and support throughout the years. I am also appreciative to my in-laws, Joe and Lisa, who’ve provided additional encouragement when I needed it the most.
To my thesis advisor Dr. Connor, thank you for your mentorship and support. I truly appreciate all that you have done for me. I also want to thank my committee members; thank you all for your comments and advice throughout my research career.
Finally, I want to thank all of the members of the Connor Laboratory for all of their contributions.
Thank you Kari for your continued support and being a great friend, for whom I would not have been able to survive without. Thank you Beth for your help with the animal studies, I greatly appreciate all of your work. Also, thank you to Oliver, Brian, Mandy, and Stephanie for your friendship and support.
1 Chapter 1 Does HFE Genotype Impact Macrophage Phenotype in Disease Process and Therapeutic Response?
1.1 Abstract
Iron is a critical cofactor involved in many important physiological processes throughout the human body. Therefore, mismanagement of iron can lead to numerous unfavorable effects, such as the generation of toxic reactive oxygen species during iron overload. Iron mismanagement is also a key component of multiple neurodegenerative diseases and cancer. One way in which iron overload occurs is through a mutation within the HFE gene. However, the mechanism by which
HFE effects these diseases has yet to be elucidated. In this review, we consider how the HFE genotype may influence the role of macrophages and microglia (the resident macrophage of the brain) in neurodegenerative disease and cancer. Macrophages appear to redistribute their iron to parenchymal cells when an HFE mutation is present, resulting in increased vulnerability to neurodegenerative diseases and cancer through macrophage mishandling of iron.
1.2 Iron
Iron is essential in the regulation of many biochemical and cellular processes within the body, including DNA synthesis, oxygen transport in blood, and energy metabolism in the mitochondria
1,2. Too much or too little iron can have detrimental effects, therefore iron is tightly regulated.
Once in the bloodstream, approximately two-thirds of transferrin-bound iron is absorbed by erythroid precursors where it becomes incorporated into hemoglobin, while the rest is circulated and utilized by muscle and other tissues including the brain3. Iron is stored primarily within the liver parenchymal cells and reticular endothelial cells, such as macrophages4. In the blood, iron is primarily bound to transferrin. Transferrin contains two binding sites for iron. Therefore transferrin can exist in three different states; free of iron (apo-transferric), one bound iron ion
2 (mono-ferric), or saturated in which two iron ions are bound (di-ferric)5-8. The majority of circulating iron is bound to transferrin in the mono-ferric9. However, in iron overload states such as hemochromatosis, majority of transferrin is saturated 10.
Iron bound to transferrin (di-ferric transferrin) enters cells by binding to transferrin receptor-1
(TfR-1) on the cell surface 11-13. TfR-1 is a homodimer, which becomes endocytosed upon transferrin binding. Subsequently, a V-type proton ATPase causes the release of iron from transferrin within the endosome at pH of 5.5 14. Iron is then reduced from ferric (Fe3+) to ferrous iron (Fe2+) by the six transmembrane epithelial antigen of the prostate-3 (STEAP-3), allowing it to leave the endosome through DMT-115. Once in the cytoplasm, iron can be stored within ferritin
16,17. Ferritin is composed of 24 polypeptide subunits that can store up to 4,500 ferric atoms. It consists of a mixture of heavy (H-ferritin) and light (L-ferritin) subunits17. H-ferritin induces oxidation of ferrous to ferric iron, leading to a decrease in reactive oxygen species (ROS) formation18. L-ferritin synergistically works with H-ferritin to oxidize ferrous iron by assisting in the mineralization of ferric iron19. In some instances, when cellular iron is increased more than normal, ferritin can be converted into hemosiderin, another iron storage complex that only differs based upon their physical characterizations20. Iron release from ferritin occurs through either proteosomal or lysosomal degradation21,22.
1.2.1 Iron Regulation
As stated previously, iron plays an important role in cell growth and metabolism. Without proper regulation, iron can promote the formation of free radicals, which cause oxidative damage and can be toxic to cells 23. Regulation of iron at the cellular level is controlled by uptake and storage24. Regulation of the uptake and storage system at the cellular level begins with the iron regulatory protein/iron regulatory element (IRP/IRE) system, which either suppresses or activates
3 transcription of the proteins involved in iron homeostasis. IRPs, located in the cytoplasm, bind to
IREs found on untranslated regions (UTR) of mRNA on both their 3’ and 5’ ends 25,26. Low cellular iron levels result in translation of TfR mRNA by IRPs binding to the 3’ UTR of TfR and preventing degradation by stabilization of TfR mRNA 27-30. Ultimately, the binding of IRPs results in decreased iron storage and increased iron import. Conversely, when there are high levels of iron within the cell, the IRPs fail to bind to the 5’ ferritin UTR promoting translation of ferritin mRNA. Thus, this results in an increase in ferritin and iron storage and a decrease in TfR expression and subsequent iron uptake30.
At the systemic level, regulation of iron homeostasis occurs mainly by controlling absorption from the gut, by the circulating peptide hepcidin. Hepcidin is a 25 amino acid peptide secreted from hepatocytes, which binds to ferroportin resulting in its internalization and degradation, preventing iron export into the circulation31. When levels of iron are high within the blood, the liver increases the production of hepcidin, inhibiting iron release from enterocytes, macrophages, and other cells 32. These mechanisms are essential in maintaining iron homeostasis. Dysregulation of these key mechanisms lead to iron related diseases such iron deficiency or iron overload.
1.3 Hemochromatosis
1.3.1 Discovery
Hemochromatosis is an iron overload disorder that results in increased total body iron, primarily within the parenchymal cells. If not treated the iron overload can lead to organ dysfunction and death 33. The first studies describing hemochromatosis were in 1865 by Armand Trousseau, where he observed his diabetic patients had bronze pigmented skin, cirrhosis of the liver and pancreatic fibrosis34. However, there was no pathological connection at this time to iron. It wasn’t until 1889 that Friedrich Daniel von Recklinghausen associated these signs with iron. Upon sectioning and
4 staining livers from deceased patients, he found increased iron pigmentation within liver cells.
Believing that the iron was derived from blood, he in turn named the disorder hemochromatosis35.
The cause of hemochromatosis was still unknown until Joseph Sheldon saw an increase in iron in all body regions including blood, brain, and colon. This, along with the increased frequency of hemochromatosis in males and familial incidence resulted in his belief that hemochromatosis was hereditary in nature36,37. Despite these early findings, the actual gene that causes hemochromatosis was not discovered until 1975. Simon discovered an association between hereditary hemochromatosis and HLA-A3 allele found on chromosome 6p 38,39. It was not until
20 years later that the actual gene, HLA-H, was discovered by Feder and colleagues40. It was found within the major histocompatibility complex (MHC), located specifically on chromosome
6p21.3. From these studies two missense mutations were identified within this gene: C282Y and
H63D40. Later HLA-H was renamed to HFE to avoid confusion with a previously found gene 41.
Although the HFE gene has been found to be directly associated with hemochromatosis, the precise mechanisms and functions of HFE are yet to be elucidated. The current state of knowledge for HFE function will be discussed later in the chapter.
1.3.2 Types
There are four types of hemochromatosis, each resulting in increased iron overload, by a mechanism unique to each. Hereditary hemochromatosis is type I hemochromatosis that results from a mutation within the HFE gene40. Type II hemochromatosis or juvenile hemochromatosis is a resultant of a mutation within the HAMP or HJV gene leading to misregulation of hepcidin 42,43.
Type III occurs due to a mutation within the transferrin receptor 2 gene, which is thought to contribute to the regulation of hepcidin44. Finally, type IV is an uncommon disorder with a mutation within the gene that encodes ferroportin synthesis45. Interestingly, the cause of
5 hemochromatosis, despite which type, seems to be associated with low or inadequate levels of hepcidin.
1.4 HFE
1.4.1 Structure
The HFE gene encodes the HFE protein, a type I transmembrane glycoprotein resembling class I
40 MHC molecules that associates with β2-microglobulin (β2M) molecule . It contains an α chain broken down into three domains, two of which bend to form a peptide binding groove. Despite
MHC molecules classically being able to bind to peptides, HFE lacks this feature because the α1
46 and α2 domains are too close to one another, preventing peptide binding . Instead, HFE binds to
46-48 TfR1 at a neutral pH through the α1 and α2 domains . HFE and TfR1 form a complex within the endoplasmic reticulum and are transported together through the Golgi complex to the cell surface. Without TfR1, HFE is unable to be expressed on the cellular surface49. HFE forms a complex with TfR1 by binding to one or both of the binding sites, resulting in reduced affinity of
50 TfR1 for diferric transferrin . In addition, the α2 and α3 domains contain disulfide bridges that help stabilize HFE’s tertiary structure and specifically the disulfide bridge of α3 helps interact
40,51 with β2M as seen in other class I MHC molecules .
1.4.2 Function
Due to HFE’s ability to bind to TfR1, HFE indirectly regulates iron homeostasis 46,48. The molecular mechanism by which HFE affects iron homeostasis is still not well understood. One proposal states binding of wildtype HFE to TfR1 limits the amount of iron imported into a cell because only one molecule of iron bound transferrin can bind to TfR1. Additionally, studies have shown that in the presence of HFE, cells express reduced levels of Fe2-Tf uptake, decreased ferritin levels, and an increase in TfR149,52-55. Therefore, a mutation within the gene results in
6 parenchymal cells accumulating excess iron. Recent studies propose that HFE also interacts with another transferrin receptor found primarily in the liver, TfR2, after its dissociation from TfR156.
However, the interaction between HFE and TFR2 is at the α3 domain and does not compete with transferrin 57. This interaction is understood to regulate hepcidin expression through the bone morphogenetic protein (BMP) pathway; however, the underlying mechanism remains unidentified. Briefly, HFE may bind to the BMP type I receptor ALK3 or may bind to the BMP co-receptor, hemojuvelin (HJV), along with TfR2 resulting in phosphorylation of the Smad 1/5/8 proteins 58-60. In turn, these proteins are transported to the cell’s nucleus along with the protein
Smad 4, resulting in hepcidin transcription61. Ultimately, hepcidin production is increased and will limit the amount of iron released by its previously described interaction with ferroportin 31.
1.4.3 Polymorphisms
There are two main polymorphisms of HFE, C282Y, H63D, and one less commonly studied polymorphism, S65C. C282Y and H63D polymorphisms were elucidated through the discovery of the HFE gene and account for over 90% of the hemochromatosis alleles 40,62. Not all cases of hemochromatosis are caused by these two missenses, which led to the discovery of the alternative
S65C mutation 62.
C282Y results from a substitution of a tyrosine with a cysteine at position 282. It is found to be homozygous in 80-90% of patients with hemochromatosis, the most common mutation within the hemochromatosis population 40,47,63,64. Unlike wildtype HFE, the C282Y mutation does not interact with the β2M molecule as it is localized intracellularly and unable to migrate to the cellular membrane. Thus, it fails to interact with TfR1, resulting in increased iron import as more transferrin binding sites are available 65-67. Additionally, the C282Y mutation is able to interact with ALK3, a receptor for BMP type 1(BMP1), similar to wildtype HFE. This interaction results
7 in the inhibition of hepcidin expression even though ALK3 ubiquitination was not prevented.
Despite not inhibiting ALK3 ubiquination, cell surface expression of ALK3 was not increased 59.
The H63D mutation is caused by a replacement of histidine for aspartic acid at position 6340. It is the second most common mutation found in hemochromatosis patients with an allele frequency between 15-20% 68,69. Of note, H63D is the most common mutation within the Caucasian population even though it is not the most prevalent mutation in hemochromatosis patients 40,68,70.
A number of studies have examined the H63D mutation in neurodegenerative diseases which will be discussed later. Unlike the C282Y mutation, H63D is able to migrate to the cellular membrane through its association with β2M and binds to TfR1 with the same affinity as wildtype
HFE66,71.This may affect the interaction of HFE with associating proteins or ligands, nevertheless the mechanism is poorly understood47. However, it is known that H63D is able to interact with
ALK3, and like the C282Y mutation it does not induce hepcidin expression59. Conversely, H63D causes ubiquination of ALK3 and thus results in decreased ALK3 protein levels.
The S65C mutation occurs with a lower frequency in hereditary hemochromatosis than C282Y and H63D mutations, affecting around 2% of the hemochromatosis population 62,72. S65C is the resultant of a serine substitution for a cysteine on exon two, linked to HLA-3272. These patients rarely are homozygous and instead have one chromosome without a mutation, or in combination with the C282Y or H63D mutation. This may account for a milder iron overload in which patients present with normal transferrin saturation and the C282Y and H63D mutation may be needed to produce symptoms of hemochromatosis62,73. However, elevated transferrin saturation levels have been reported in S65C mutations, but it did not cause hepatic fibrosis or cirrhosis, the result of chronic iron overload 62,74.
8 1.4.4 HFE and Neurodegenerative Diseases
The relationship between HFE status and the brain has been fraught with misinformation and misquoting of early literature. The prevailing paradigm was that blood brain barrier kept the amount of iron static and protected the brain from increased iron accumulation. The initial autopsy reports on brains of patients with hemochromatosis reported brain iron accumulation not only in those areas unprotected by the blood-brain-barrier (BBB), but also in regions protected by the BBB 37,75. Supporting this, MRIs have shown increased iron accumulation within various brain regions76,77, and animal models of the HFE H63D mutation, unlike HFE knockout models, also report increased iron in the brain78,79.
Mutations in HFE are associated with increased susceptibility to neurodegenerative diseases 80, and also appear to disrupt other processes such as myelination and cholesterol metabolism79.
Loss of iron homeostasis leads to increased oxidative stress and neuronal death which are hallmarks of neurodegenerative disease 78,81,82. The incidence of the HFE gene variants have been linked to Alzheimer’s disease (AD), Amyotrophic Lateral Sclerosis (ALS), and Parkinson’s disease (PD) 83-88, indicating a general critical contribution of the HFE protein in maintaining the brain in a healthy state. However, there are also studies have reported that the HFE gene variants may be protective against neurodegenerative diseases and others have found no association between the gene and diseases86,89-92.The lack of agreement is likely associated with gene- environment interaction because, as pointed out earlier, iron is only obtained by the body via the diet, therefore variations in dietary access to iron are likely to impact the extent of HFE genotype effects on disease. Given the prevalence of the HFE mutations, especially the H63D variant in the general population it is likely that the presence of this mutation impacts disease processes and therapeutic response. For a more detailed review of HFE mutations and neurodegenerative disease, refer to Nandar and Connor, 201182. In this review we will focus on microglial
9 responses in various diseases and how their response may be impacted by the HFE genotype.
Microglia are the resident macrophages in the brain and thus the impact of the HFE genotype on microglial function should be similar to that of macrophages in other organs. Microglial activation is reported as part of all neurodegenerative diseases93.
1.4.4.1 Alzheimer’s Disease
In general, accumulation of iron within neurofibrillary tangles, senile plaques, and oxidative stress are pathological hallmarks of AD 94-96. The oxidative stress in AD can cause degeneration of microglia, resulting in a limitation of neuroprotection97. In the presence of the HFE mutation, microglia could increase iron export which can lead to an increase in oxidative damage. Just like macrophages, microglia are phagocytic and should be able to clear amyloid-β (Aβ) plaques making them in sense neuroprotective. However, with age the function of microglia diminishes98, which would contribute to the accumulation of Aβ plaques. Furthermore, cell culture studies have shown that even though microglia may initially phagocytose Aβ plaques this phenomenon does not continue and the amount of degradation is limited 99-101. In cell culture models, expression of
H63D mutation within neuroblastoma cells caused an increase in tau phosphorylation and decreased levels of Prolylpeptidyl isomerase-1, which are involved in the production of neurofibrillary tangles and phosphorylation of amyloid precursor protein (APP) and tau, respectively 102. The presence of the H63D variant increases cellular sensitivity to AB toxicity.
Furthermore, HFE is increased within the area of amyloid beta plaques of AD brains 103.
1.4.4.2 Amyotrophic Lateral Sclerosis
In ALS, the HFE mutation is associated with a 4-fold increase in risk for ALS 83,84 and present in up to 30% of cases 83,84,104,105. Neuronal cell culture models and a mouse model that carries the
H63D gene variant have shown increased levels of oxidative stress, glutamate, and endoplasmic
10 reticulum stress which are consistent findings in models of ALS 78,106-108. Additionally, when combined with the SOD1 mutation, the H63D gene variant accelerates disease progression and shorten life expectancy in the mouse model. Furthermore, microglia within this model contain increased levels of L-ferritin and overall had more activated microglia than wildtype counterpart
109. The increase in L-ferritin could indicate increased levels of iron and pro-inflammatory cytokines and ultimately lead to increase in oxidative stress, microglia toxicity, and accelerated disease progression110. Just as with Alzheimer’s Disease, microglia may have a protective effect of motor neurons during early stages of ALS due to low levels of microglia present in spinal cord111. However, in later stages there are significantly increased levels of activated microglia which may lead to accelerated disease progression through their secretion of pro-inflammatory cytokines and reactive oxygen species 112,113.
1.4.4.3 Parkinson’s Disease
There have been many clinical studies looking at the incidence between HFE and PD, however the results are conflicting. Multiple studies have shown an increase in frequency between the
HFE mutation, C282Y and PD87,88, however others have shown no association of HFE genotypes and PD114-116. In PD, the HFE gene variant is associated with an increase in α-synuclein, which enhances the generation of Lewy bodies a pathological hallmark of PD 117,118. Just as in AD and
ALS, PD has an increase in activated microglia. This activation in PD however can come from α- synuclein through subsequent activation of toll-like receptor119. Furthermore, characterization of
PD is the loss of dopaminergic neurons which may be mediated through the production of inflammation and ROS from activated microglia120.
11 1.4.5 HFE and Cancer
HFE gene variants have been linked to an increased risk for certain cancers, once again due to its effect on iron metabolism but also its contribution to immune function121. Despite the fact that studies looking at the relationship between HFE and cancer have tended to focus on iron overload in the liver and hepatic cancer, associations with many other cancers have been found. Several studies have found HFE mutations can be cyto-protective, suggesting a tumor suppressive role
96,108,122-125. Opposing this, the H63D mutation has also been shown to be significantly increased in breast cancer, hepatocellular carcinoma, and gliomas 126-129, suggesting the loss of function of
HFE increases tumor progression. It was also found that patients expressing this mutation had an increased risk of pediatric acute lymphocytic leukemia (ALL), colorectal cancer, gastric cancer, and hepatocellular carcinoma 130-134. Similarly, increased C282Y mutations have been found in breast cancer, colorectal cancer, hepatocellular carcinoma and pediatric ALL 128,130,135-138.
Furthermore, those with a C282Y have increased risk for developing pediatric ALL, breast cancer, colorectal cancer, hepatocellular carcinoma, and ovarian cancer 130,132,139-143. However, there are also studies that have reported neither an association of the HFE variants to patients with cancer, nor their risk for developing cancer 134,144-147. The differences in these studies may be attributed to their differences in methodologies, such as a lack of controls and limited patient population but also the question of gene and environment interaction mentioned earlier.
A key question lacking from these studies is how macrophages are influencing particularly solid tumors. As discussed previously, macrophages can be activated and polarized into two different populations, of which M2 (alternatively-activated macrophages) are most commonly associated with tumor promotion. M2 macrophages play a key role in tumor growth and progression through an increase in iron release that in turn promotes cellular growth and DNA replication. Tumor associated macrophages (TAMs) secrete components that aide in the angiogenesis, recruitment of
12 additional macrophages, and lymphangiogensis148,149 These all allow the tumors to become highly vascularized and increase their growth and migration. In those reports where the HFE mutation has been shown be protective one wonders if the tumoricidal activity of macrophages was activated differently. Weston et al. showed that upon incubating conditioned media from H67D
(H63D mouse homolog) macrophages on B16F10 cells, there was significantly smaller tumor growths compared to B16F10 cells incubated with wildtype macrophages 150. Corresponding to these results, other studies have shown TAMs to secrete pro-inflammatory cytokines which can destroy tumor cells, and initiate a T-cell anti-tumor response151.
The ability of macrophages to take up iron is being explored, for imaging of brain tumors.
Superparamagnetic iron oxide nanoparticles are under consideration as image enhancing agents for CNS brain tumors by taking advantage of macrophages and their aggressive iron uptake in the tumor microenvironment152. Given the utilization of iron by the macrophages for pro- inflammatory activity, some caution should be given until we learn how iron loading macrophages in tumors may impact the tumor or how the HFE genotype affects the phenotype of tumor associated macrophages.
1.4.6 HFE Animal Models
A number of mouse models and cell lines have been created in an attempt to model the impact of
HFE gene variants and identify the mechanisms involved. The HFE gene found in mice is structurally similar to the human HFE gene and contains a 66% conserved sequence between the two species152. Different mouse models disrupting the HFE gene include, HFE knockout, C294Y knockin (mouse homolog for C282Y), and H67D knockin (mouse homolog for H63D) 78,153-156.
All of the models containing a knockout of HFE result in significant iron overload compared to the control mice, in addition to increased transferrin saturation and plasma levels. Predominately,
13 iron accumulation was found in liver localized in the hepatocytes. There was not a significant change in iron levels found in other organs such as the spleen, kidneys, and heart 153-155. In comparison to HFE null mice, HFE knockin mice also have increased iron loading however it is not as significant as the null mice 155,156. Furthermore, the H67D mutation did not induce as severe iron loading as C294Y, however the amount of iron was significantly greater than wildtype mice 156. This observation could be relevant to the clinical data where C282Y is much more common in hemochromatosis than the H63D mutation 40,157. Few studies have examined changes in the brain with the HFE gene variants. Nandar et al. 109 found their H67D knockin model to have increased levels of ferritin, HFE, and oxidative stress along with decreased transferrin. Furthermore, Ali-Rahmani 79 found increased caspase-3 levels, decreased synaptic proteins and lower spatial memory and recognition in the same H67D knockin model. These mice showed structural changes in the brain similar to that seen in aging humans with an H63D mutation158. These results clearly establish that the HFE genotype is relevant to human neurological disease and provide models for studying the role of macrophages and microglia in iron management in disease.
1.5 Macrophages
Macrophages are components of the mononuclear phagocytic system, and critical regulators of the immune system. They are versatile cells that provide host defense, destroying invading pathogens and initiating an immune response. Understanding the function of macrophages is critical because despite the macrophage’s ability to protect the host, they can also contribute to inflammatory and degenerative diseases.
14 1.5.1 Macrophages and Iron Metabolism
Macrophages are a key component in the regulation of iron homeostasis as they recycle damaged erythrocytes and other senescent cells that can be used in the production of new erythrocytes and iron management proteins (IMPs) 153. Similar to other cell types, iron is able to enter macrophages through transferrin receptor-1 when bound to transferrin14,159. Macrophages also acquire iron through phagocytosis and endocytosis, consistent with their primary function of eliminating cellular debris and foreign particles. Within the macrophage, heme oxygenase-1(HO-
1) releases iron through degradation of phagocytosed stressed and damaged erythrocytes, and endocytosed heme and hemoglobin160. Iron is then exported into the cytoplasm via DMT-1 and
NRAMP-1 in phagolysomes161; the latter protein specific to macrophages. Once in the cytoplasm, iron can be stored within ferritin17. If the need for iron in the body is great, iron can bypass ferritin storage and be exported immediately through ferroportin once oxidized to ferric iron162. Ferritin can also be released from macrophages and microglia163. In the latter cell type, ferritin can be a trophic influence on oligodendrocytes164. The release of ferritin may occur through a lysosomal secretory pathway165. As macrophages and microglia are activated during inflammation, they release ferritin resulting in the increased serum ferritin levels found during inflammatory states166. Secretion of iron through ferritin is underrepresented pathway, which may be key in the regulation and secretion and source of iron.
1.5.2 Macrophage Polarization
Macrophages also have a role in cell-mediated immunity and wound repair, and both are regulated through iron homeostasis167-169. Macrophages possess the unique ability to become polarized in response to different environmental stimuli170. Generally, polarized macrophages can be classified into two groups; classically activated (M1) or alternatively activated (M2) macrophages. M1 macrophages are activated for cell mediated immune response to provide host
15 defense. During stress or infections, natural killer cells produce interferon gamma (IFNγ) that prime macrophages into classically activated macrophages, resulting in their secretion of pro- inflammatory cytokines such as IL-6, IL-1, and TNF-α, and other innate immune mediators that help defend and kill off pathogens and infections171,172. M1 macrophages contain higher levels of ferritin and lower transferrin and ferroportin levels 173. These findings may be attributed to their function, in which low levels of iron are exported to limit the growth of an infection and protect against oxidative damage. In addition, macrophages may be holding onto iron by increased hepcidin levels that occurs when IL-6 binds to its receptor resulting in activation Janus kinases that in turn phosphorylate STAT3. Subsequently, STAT3 is transported to the nucleus where it induces the transcription of many genes, including hepcidin 174. This increase in hepcidin production attributes to the decreased levels of ferroportin, transferrin receptor and iron export, and increased ferritin levels 173. This elegant system for limiting the bioavailability of iron during infections was introduced in 1984 as the iron withholding defense175.
Alternatively activated M2 macrophages share some similarities to M1 macrophages. However, they are considered “healers” compared to the M1 “killer” macrophages. They are recruited in wound healing and debris scavenging, but have also play a role in promoting tumorigenesis 176-181.
The production of IL-4 from injured tissues activates M2 macrophages and causes them to secrete precursors and components of extracellular matrix 176,177. M2 macrophages play a pro-tumor role by promotion angiogenesis. In fact, tumor associated macrophages are a subpopulation of M2181.
In comparison to M1 macrophages, M2 macrophages have lower levels of ferritin, but higher levels of transferrin receptor and ferroportin. As previously stated, M1 macrophages secrete pro- inflammatory cytokines that attribute to their IMP characterization, whereas M2 macrophages secrete low levels of pro-inflammatory cytokines but high levels of anti-inflammatory cytokines
16 (IL-13). These low levels of pro-inflammatory cytokines subsequently result in a decrease in the production of hepcidin, thus leading to decreased levels of ferroportin 173.
1.5.3 Macrophages and HFE
Reticuloendothelial cells, primarily macrophages, are reportedly iron poor in iron overload disorders such as hereditary hemochromatosis, despite the high levels of iron within liver, spleen, and bone marrow 182,183. Under normal conditions, the amount of iron within parenchymal cells is similar to that of reticuloendothelial cells. The mechanism behind macrophages abnormal iron levels is poorly understood, but Montosi et al. 184 shows that HFE functions differently in macrophages when compared to parenchymal cells. Wild-type HFE macrophages have increased iron and ferritin expression following exposure to radiolabeled Fe-Tf. In addition, there were also decreased TfR-1 levels correlating with the increased levels of iron and ferritin184.
Drakesmith et al.185 also found that wild-type HFE can increase iron levels within macrophages, through inhibition of iron export not enhanced iron uptake. Comparing the wild-type HFE macrophages to those transfected with the H63D mutation, the wild-type HFE was associated with an increased inhibition of iron release. Furthermore, it is proposed there is a homeostatic mechanism linking HFE binding to transferrin receptor-1 and ferroportin. When transferrin saturation levels are low, HFE remains bound to TfR-1, not ferroportin. Therefore, iron export is increased from macrophages and decreases intracellular iron concentration. The opposite occurs when transferrin saturation is high; HFE is blocked from binding to TfR-1 and instead binds to ferroportin, sequestering iron within the cell186. The loss of homeostasis between these two mechanisms decreases the capacity of iron storage within macrophages. Contrast to these results,
Waheed et al. 66showed an increase in iron uptake in Chinese hamster ovary cells transfected with wild-type HFE. Therefore, mutations with in the HFE gene would prevent enhanced iron uptake.
17 Overall, the mechanism regarding macrophages as a function of HFE genotype is still unclear as there are many limitations to the studies presented. The previous studies describe looked only at the role of HFE in wildtype macrophages, however this still does not answer the question of why the mutated macrophages are iron deficient. Furthermore, the macrophages obtained by
Drakesmith consisted of a monocytic cell line transfected with HFE, which may not accurately reflect in-situ process, leaving the role of HFE in macrophage iron management still undiscovered.
1.6 Conclusions
A key cellular component of neurodegenerative disease and cancer is the macrophage/microglia, and iron is a determining factor in the function of these cells; not only in the phenotype of these cells but in the manner in which they influence disease process. The HFE genotype alters the incidence and the disease course in a number of diseases. In this review we propose that altered macrophage function as a result of HFE genotype could underlie the impact of HFE genotype on disease pathogenesis.
1.7 Acknowledgements
Sections 1.1 through 1.6 have previously been published: Nixon AM and Connor JR. Does HFE
Genotype Impact Macrophage Phenotype in Disease Process and Therapeutic Response? In:
White AR, Aschner M, Costa LG, Bush AI, eds., Biometals in Neurodegenerative Diseases:
Mechanisms and Therapeutics, San Diego: Academic Press, 2017: 51-66. © 2017 Elsevier B.V.
All rights reserved. Inclusion of this material in the dissertation is in accordance with rights retained by authors described in the publisher’s copyright policy
18 1.8 References
1. Wang J, Pantopoulos K. Regulation of cellular iron metabolism. The Biochemical
journal. 2011;434(3):365-381.
2. Andrews NC. Disorders of Iron Metabolism. New England Journal of Medicine.
1999;341(26):1986-1995.
3. Andrews NC. Disorders of iron metabolism. N Engl J Med. 1999;341(26):1986-1995.
4. Bothwell TH, Charlton RW, Cook JD, Finch CA. Iron metabolism in man. Oxford:
Blackwell Scientific Publications.; 1979.
5. Holmberg CG, Laurell CB. Investigations in serum copper. I. Nature of serum copper and
its relation to the iron-binding protein in human serum. Acta Chemica Scandinavica.
1947;1:944–950.
6. Laurell CB, Ingelman B. The iron-binding protein of swine serum. Acta Chemica
Scandinavica. 1947;1:770–776.
7. Harris DC. Functional equivalence of iron bound to human transferrin at low pH or high
pH. Biochimica et Biophysica Acta (BBA) - General Subjects. 1977;496(2):563-565.
8. Huebers HA, Csiba E, Huebers E, Finch CA. Competitive advantage of diferric
transferrin in delivering iron to reticulocytes. Proceedings of the National Academy of
Sciences of the United States of America. 1983;80(1):300-304.
9. Fletcher J. Iron transport in the blood. Proceedings of the Royal Society of Medicine.
1970;63(12):1216-1218.
10. Lee G, J F, Lukens J, et al. Hemochromatosis. In: Lee G, J F, J L, F P, JP G, GM R, eds.
Wintrobe’s Clinical Hematology. Vol 1. 10 ed. Baltimore, MD: Williams and Wilkens;
1999:1056-1070.
19 11. Jandl JH, Inman JK, Simmons RL, Allen DW. Transfer of iron from serum iron-binding
protein to human reticulocytes. The Journal of Clinical Investigation. 1959;38(1):161-
185.
12. Garrett NE, Burriss Garrett RJ, Archdeacon JW. Solubilization and chromatography of
iron-binding compounds from reticulocyte stroma. Biochemical and Biophysical
Research Communications. 1973;52(2):466-474.
13. Leibman A, Aisen P. Transferrin receptor of the rabbit reticulocyte. Biochemistry.
1977;16(7):1268-1272.
14. Dautry-Varsat A, Ciechanover A, Lodish HF. pH and the recycling of transferrin during
receptor-mediated endocytosis. Proceedings of the National Academy of Sciences.
1983;80(8):2258-2262.
15. Ohgami RS, Campagna DR, Greer EL, et al. Identification of a ferrireductase required for
efficient transferrin-dependent iron uptake in erythroid cells. Nat Genet.
2005;37(11):1264-1269.
16. Picard V, Govoni G, Jabado N, Gros P. Nramp 2 (DCT1/DMT1) expressed at the plasma
membrane transports iron and other divalent cations into a calcein-accessible cytoplasmic
pool. The Journal of biological chemistry. 2000;275(46):35738-35745.
17. Harrison PM, Arosio P. The ferritins: molecular properties, iron storage function and
cellular regulation. Biochimica et biophysica acta. 1996;1275(3):161-203.
18. Levi S, Luzzago A, Cesareni G, et al. Mechanism of ferritin iron uptake: activity of the
H-chain and deletion mapping of the ferro-oxidase site. A study of iron uptake and ferro-
oxidase activity of human liver, recombinant H-chain ferritins, and of two H-chain
deletion mutants. Journal of Biological Chemistry. 1988;263(34):18086-18092.
20 19. Levi S, Santambrogio P, Cozzi A, et al. The Role of the L-Chain in Ferritin Iron
Incorporation: Studies of Homo and Heteropolymers. Journal of Molecular Biology.
1994;238(5):649-654.
20. Shoden A, Gabrio BW, Finch CA. The relationship between ferritin and hemosiderin in
rabbits and man. The Journal of biological chemistry. 1953;204(2):823-830.
21. Larson JA, Howie HL, So M. Neisseria meningitidis accelerates ferritin degradation in
host epithelial cells to yield an essential iron source. Molecular Microbiology.
2004;53(3):807-820.
22. Mehlhase J, Sandig G, Pantopoulos K, Grune T. Oxidation-induced ferritin turnover in
microglial cells: role of proteasome. Free radical biology & medicine. 2005;38(2):276-
285.
23. Fenton H. Oxidation of tartaric acid in presence of iron. J Chem Soc. 1894;65:899-903
24. Green R, Charlton R, Seftel H, et al. Body iron excretion in man: A collaborative study.
The American Journal of Medicine. 1968;45(3):336-353.
25. Hentze MW, Caughman SW, Rouault TA, et al. Identification of the Iron-Responsive
Element for the Translational Regulation of Human Ferritin mRNA. Science.
1987;238(4833):1570-1573.
26. Hentze MW, Caughman SW, Casey JL, et al. A model for the structure and functions of
iron-responsive elements. Gene. 1988;72(1–2):201-208.
27. Aziz N, Munro HN. Iron regulates ferritin mRNA translation through a segment of its 5'
untranslated region. Proc Natl Acad Sci U S A. 1987;84(23):8478-8482.
28. Leibold EA, Munro HN. Cytoplasmic protein binds in vitro to a highly conserved
sequence in the 5' untranslated region of ferritin heavy- and light-subunit mRNAs.
Proceedings of the National Academy of Sciences of the United States of America.
1988;85(7):2171-2175.
21 29. Casey JL, Jeso BD, Rao K, Klausner RD, Harford JB. Two Genetic Loci Participate in
the Regulation by Iron of the Gene for the Human Transferrin Receptor. Proceedings of
the National Academy of Sciences of the United States of America. 1988;85(6):1787-
1791.
30. Casey JL, Hentze MW, Koeller DM, et al. Iron-Responsive Elements: Regulatory RNA
Sequences that Control mRNA Levels and Translation. Science. 1988;240(4854):924-
928.
31. Nemeth E, Tuttle MS, Powelson J, et al. Hepcidin Regulates Cellular Iron Efflux by
Binding to Ferroportin and Inducing Its Internalization. Science. 2004;306(5704):2090-
2093.
32. Ramos E, Kautz L, Rodriguez R, et al. Evidence for distinct pathways of hepcidin
regulation by acute and chronic iron loading in mice. Hepatology. 2011;53(4):1333-1341.
33. Piperno A. Classification and diagnosis of iron overload. Haematologica.
1998;83(5):447-455.
34. Trousseau A. Glycosurie; diabete sucre. Clinique Med de l’Hotel de Paris. 1865;2:663–
698.
35. von Recklinghausen F. Uber Hamochromatose. Tagebl Versamml Natur Artze
Heidelberg. 1889;62:324-235.
36. Ramage H, Sheldon J. Haemochromatosis. Quarterly Journal of Medicine. 1934;4:121-
129.
37. Sheldon J. Haemochromatosis. Oxford University Press; 1935.
38. Simon M, Pawlotsky Y, Bourel M, Fauchet R, Genetet B. Hemochromatose idiopathique:
maladie associee a l'anligine tissulaire HLA 3? La Nouvelle presse medicale.
1975;4(19):1432.
22 39. Simon M, Bourel M, Fauchet R, Genetet B. Association of HLA-A3 and HLA-B14
antigens with idiopathic haemochromatosis. Gut. 1976;17(5):332-334.
40. Feder JN, Gnirke A, Thomas W, et al. A novel MHC class I-like gene is mutated in
patients with hereditary haemochromatosis. Nat Genet. 1996;13(4):399-408.
41. Bodmer JG, Marsh SGE, Albert ED, et al. Nomenclature for Factors of the HLA System,
1996. Human Immunology. 1997;53(1):98-128.
42. Roetto A, Papanikolaou G, Politou M, et al. Mutant antimicrobial peptide hepcidin is
associated with severe juvenile hemochromatosis. Nat Genet. 2003;33(1):21-22.
43. Papanikolaou G, Samuels ME, Ludwig EH, et al. Mutations in HFE2 cause iron overload
in chromosome 1q-linked juvenile hemochromatosis. Nat Genet. 2004;36(1):77-82.
44. Nemeth E, Roetto A, Garozzo G, Ganz T, Camaschella C. Hepcidin is decreased in TFR2
hemochromatosis. Blood. 2004;105(4):1803-1806.
45. De Domenico I, Ward DM, Nemeth E, et al. The molecular basis of ferroportin-linked
hemochromatosis. Proceedings of the National Academy of Sciences of the United States
of America. 2005;102(25):8955-8960.
46. Lebrón JA, Bennett MJ, Vaughn DE, et al. Crystal Structure of the Hemochromatosis
Protein HFE and Characterization of Its Interaction with Transferrin Receptor. Cell.
1998;93(1):111-123.
47. Feder JN, Penny DM, Irrinki A, et al. The hemochromatosis gene product complexes
with the transferrin receptor and lowers its affinity for ligand binding. Proceedings of the
National Academy of Sciences. 1998;95(4):1472-1477.
48. Bennett MJ, Lebron JA, Bjorkman PJ. Crystal structure of the hereditary
haemochromatosis protein HFE complexed with transferrin receptor. Nature.
2000;403(6765):46-53.
23 49. Salter-Cid L, Brunmark A, Li Y, et al. Transferrin receptor is negatively modulated by
the hemochromatosis protein HFE: Implications for cellular iron homeostasis.
Proceedings of the National Academy of Sciences of the United States of America.
1999;96(10):5434-5439.
50. Giannetti AM, Björkman PJ. HFE and Transferrin Directly Compete for Transferrin
Receptor in Solution and at the Cell Surface. Journal of Biological Chemistry.
2004;279(24):25866-25875.
51. Bjorkman PJ, Parham P. Structure, Function, and Diversity of Class I Major
Histocompatibility Complex Molecules. Annual Review of Biochemistry. 1990;59(1):253-
288.
52. Roy CN, Penny DM, Feder JN, Enns CA. The Hereditary Hemochromatosis Protein,
HFE, Specifically Regulates Transferrin-mediated Iron Uptake in HeLa Cells. Journal of
Biological Chemistry. 1999;274(13):9022-9028.
53. Corsi B, Levi S, Cozzi A, et al. Overexpression of the hereditary hemochromatosis
protein, HFE, in HeLa cells induces an iron-deficient phenotype. FEBS Letters.
1999;460(1):149-152.
54. Roy CN, Carlson EJ, Anderson EL, et al. Interactions of the ectodomain of HFE with the
transferrin receptor are critical for iron homeostasis in cells. FEBS Letters.
2000;484(3):271-274.
55. Gross CN, Irrinki A, Feder JN, Enns CA. Co-trafficking of HFE, a Nonclassical Major
Histocompatibility Complex Class I Protein, with the Transferrin Receptor Implies a Role
in Intracellular Iron Regulation. Journal of Biological Chemistry. 1998;273(34):22068-
22074.
24 56. Gao J, Chen J, Kramer M, Tsukamoto H, Zhang A-S, Enns CA. Interaction of the
Hereditary Hemochromatosis Protein, HFE, with Transferrin Receptor 2 Is Required for
Transferrin-Induced Hepcidin Expression. Cell metabolism. 2009;9(3):217-227.
57. Chen J, Chloupková M, Gao J, Chapman-Arvedson TL, Enns CA. HFE Modulates
Transferrin Receptor 2 Levels in Hepatoma Cells via Interactions That Differ from
Transferrin Receptor 1-HFE Interactions. Journal of Biological Chemistry.
2007;282(51):36862-36870.
58. D’Alessio F, Hentze MW, Muckenthaler MU. The hemochromatosis proteins HFE,
TfR2, and HJV form a membrane-associated protein complex for hepcidin regulation.
Journal of Hepatology. 2012;57(5):1052-1060.
59. Wu X-g, Wang Y, Wu Q, et al. HFE interacts with the BMP type I receptor ALK3 to
regulate hepcidin expression. Blood. 2014;124(8):1335-1343.
60. Ryan JD, Ryan E, Fabre A, Lawless MW, Crowe J. Defective bone morphogenic protein
signaling underlies hepcidin deficiency in HFE hereditary hemochromatosis. Hepatology.
2010;52(4):1266-1273.
61. Heldin C-H, Miyazono K, ten Dijke P. TGF-[beta] signalling from cell membrane to
nucleus through SMAD proteins. Nature. 1997;390(6659):465-471.
62. Mura C, Raguenes O, Férec C. HFE Mutations Analysis in 711 Hemochromatosis
Probands: Evidence for S65C Implication in Mild Form of Hemochromatosis. Blood.
1999;93(8):2502-2505.
63. Jouanolle AM, Gandon G, Jezequel P, et al. Haemochromatosis and HLA-H. Nat Genet.
1996;14(3):251-252.
64. Beutler E, Gelbart T, West C, et al. Mutation Analysis in Hereditary Hemochromatosis.
Blood Cells, Molecules, and Diseases. 1996;22(2):187-194.
25 65. Feder JN, Tsuchihashi Z, Irrinki A, et al. The Hemochromatosis Founder Mutation in
HLA-H Disrupts β2-Microglobulin Interaction and Cell Surface Expression. Journal of
Biological Chemistry. 1997;272(22):14025-14028.
66. Waheed A, Parkkila S, Zhou XY, et al. Hereditary hemochromatosis: Effects of C282Y
and H63D mutations on association with β(2)-microglobulin, intracellular processing,
and cell surface expression of the HFE protein in COS-7 cells. Proceedings of the
National Academy of Sciences of the United States of America. 1997;94(23):12384-
12389.
67. Feeney GP, Worwood M. The effects of wild-type and mutant HFE expression upon
cellular iron uptake in transfected human embryonic kidney cells. Biochimica et
Biophysica Acta (BBA) - Molecular Cell Research. 2001;1538(2–3):242-251.
68. Burt MJ, George PM, Upton JD, et al. The significance of haemochromatosis gene
mutations in the general population: implications for screening. Gut. 1998;43(6):830-836.
69. Steinberg KK, Cogswell ME, Chang JC, et al. Prevalence of C282Y and H63D Mutations
in the Hemochromatosis (HFE) Gene in the United States. JAMA. 2001;285(17):2216-
2222.
70. Marshall DS, Linfert DR, Tsongalis GJ. Prevalence of the C282Y and H63D
polymorphisms in a multi-ethnic control population. International journal of molecular
medicine. 1999;4(4):389-393.
71. Lebron JA, Bjorkman PJ. The transferrin receptor binding site on HFE, the class I MHC-
related protein mutated in hereditary hemochromatosis. J Mol Biol. 1999;289(4):1109-
1118.
72. Barton JC, Sawada-Hirai R, Rothenberg BE, Acton RT. Two Novel Missense Mutations
of the HFE Gene (I105T and G93R) and Identification of the S65C Mutation in Alabama
Hemochromatosis Probands. Blood Cells, Molecules, and Diseases. 1999;25(3):147-155.
26 73. Arya N, Chakrabrati S, Hegele RA, Adams PC. HFE S65C Variant Is Not Associated
with Increased Transferrin Saturation in Voluntary Blood Donors. Blood Cells,
Molecules, and Diseases. 1999;25(6):354-357.
74. Holmström P, Marmur J, Eggertsen G, Gåfvels M, Stål P. Mild iron overload in patients
carrying the HFE S65C gene mutation: a retrospective study in patients with suspected
iron overload and healthy controls. Gut. 2002;51(5):723-730.
75. Cammermyer J. Deposition of iron in paraventricular areas of the human brain in
hemochromatosis. J Neuropathol Exp Neurol. 1947;6(2):111-127.
76. Nielsen JE, Jensen LN, Krabbe K. Hereditary haemochromatosis: a case of iron
accumulation in the basal ganglia associated with a parkinsonian syndrome. Journal of
neurology, neurosurgery, and psychiatry. 1995;59(3):318-321.
77. Berg D, Hoggenmuller U, Hofmann E, et al. The basal ganglia in haemochromatosis.
Neuroradiology. 2000;42(1):9-13.
78. Nandar W, Neely EB, Unger E, Connor JR. A mutation in the HFE gene is associated
with altered brain iron profiles and increased oxidative stress in mice. Biochimica et
Biophysica Acta (BBA) - Molecular Basis of Disease. 2013;1832(6):729-741.
79. Ali-Rahmani F, Grigson PS, Lee S, Neely E, Connor JR, Schengrund CL. H63D
mutation in hemochromatosis alters cholesterol metabolism and induces memory
impairment. Neurobiol Aging. 2014;35(6):1511.e1511-1512.
80. Salvador GA. Iron in neuronal function and dysfunction. BioFactors. 2010;36(2):103-
110.
81. Rouault TA, Cooperman S. Brain iron metabolism. Seminars in pediatric neurology.
2006;13(3):142-148.
82. Nandar W, Connor JR. HFE Gene Variants Affect Iron in the Brain. The Journal of
Nutrition. 2011;141(4):729S-739S.
27 83. Wang X-S, Lee S, Simmons Z, et al. Increased incidence of the Hfe mutation in
amyotrophic lateral sclerosis and related cellular consequences. Journal of the
Neurological Sciences. 2004;227(1):27-33.
84. Goodall EF, Greenway MJ, van Marion I, Carroll CB, Hardiman O, Morrison KE.
Association of the H63D polymorphism in the hemochromatosis gene with sporadic
ALS. Neurology. 2005;65(6):934-937.
85. Moalem S, Percy ME, Andrews DF, et al. Are hereditary hemochromatosis mutations
involved in Alzheimer disease? American journal of medical genetics. 2000;93(1):58-66.
86. Blázquez L, De Juan D, Ruiz-Martínez J, et al. Genes related to iron metabolism and
susceptibility to Alzheimer's disease in Basque population. Neurobiology of Aging.
2007;28(12):1941-1943.
87. Dekker MC, Giesbergen PC, Njajou OT, et al. Mutations in the hemochromatosis gene
(HFE), Parkinson's disease and parkinsonism. Neurosci Lett. 2003;348(2):117-119.
88. Guerreiro R, Bras J, Santana I, et al. Association of HFE common mutations with
Parkinson's disease, Alzheimer's disease and mild cognitive impairment in a Portuguese
cohort. BMC Neurology. 2006;6(1):24.
89. Buchanan DD, Silburn PA, Chalk JB, Le Couteur DG, Mellick GD. The Cys282Tyr
polymorphism in the HFE gene in Australian Parkinson's disease patients. Neuroscience
Letters. 2002;327(2):91-94.
90. Correia AP, Pinto JP, Dias V, Mascarenhas C, Almeida S, Porto G. CAT53 and HFE
alleles in Alzheimer's disease: a putative protective role of the C282Y HFE mutation.
Neurosci Lett. 2009;457(3):129-132.
91. Li M, Wang L, Wang W, Qi XL, Tang ZY. Mutations in the HFE gene and sporadic
amyotrophic lateral sclerosis risk: a meta-analysis of observational studies. Brazilian
Journal of Medical and Biological Research. 2014;47(3):215-222.
28 92. Ali-Rahmani F, Hengst JA, Connor JR, Schengrund CL. Effect of HFE Variants on
Sphingolipid Expression by SH-SY5Y Human Neuroblastoma Cells. Neurochem Res.
2011;36(9):1687-1696.
93. Lull ME, Block ML. Microglial Activation & Chronic Neurodegeneration.
Neurotherapeutics : the journal of the American Society for Experimental
NeuroTherapeutics. 2010;7(4):354-365.
94. Lovell MA, Robertson JD, Teesdale WJ, Campbell JL, Markesbery WR. Copper, iron
and zinc in Alzheimer's disease senile plaques. J Neurol Sci. 1998;158(1):47-52.
95. Good PF, Perl DP, Bierer LM, Schmeidler J. Selective accumulation of aluminum and
iron in the neurofibrillary tangles of Alzheimer's disease: a laser microprobe (LAMMA)
study. Annals of neurology. 1992;31(3):286-292.
96. Smith MA, Harris PLR, Sayre LM, Perry G. Iron accumulation in Alzheimer disease is a
source of redox-generated free radicals. Proceedings of the National Academy of Sciences
of the United States of America. 1997;94(18):9866-9868.
97. Lopes KO, Sparks DL, Streit WJ. Microglial dystrophy in the aged and Alzheimer's
disease brain is associated with ferritin immunoreactivity. Glia. 2008;56(10):1048-1060.
98. Streit WJ, Sammons NW, Kuhns AJ, Sparks DL. Dystrophic microglia in the aging
human brain. Glia. 2004;45(2):208-212.
99. Paresce DM, Chung H, Maxfield FR. Slow degradation of aggregates of the Alzheimer's
disease amyloid beta-protein by microglial cells. J Biol Chem. 1997;272(46):29390-
29397.
100. Chung H, Brazil MI, Soe TT, Maxfield FR. Uptake, degradation, and release of fibrillar
and soluble forms of Alzheimer's amyloid beta-peptide by microglial cells. J Biol Chem.
1999;274(45):32301-32308.
29 101. Majumdar A, Chung H, Dolios G, et al. Degradation of fibrillar forms of Alzheimer's
amyloid beta-peptide by macrophages. Neurobiol Aging. 2008;29(5):707-715.
102. Hall EC, 2nd, Lee SY, Simmons Z, Neely EB, Nandar W, Connor JR. Prolyl-peptidyl
isomerase, Pin1, phosphorylation is compromised in association with the expression of
the HFE polymorphic allele, H63D. Biochimica et biophysica acta. 2010;1802(4):389-
395.
103. Connor JR, Snyder BS, Beard JL, Fine RE, Mufson EJ. Regional distribution of iron and
iron-regulatory proteins in the brain in aging and Alzheimer's disease. Journal of
neuroscience research. 1992;31(2):327-335.
104. Restagno G, Lombardo F, Ghiglione P, et al. HFE H63D polymorphism is increased in
patients with amyotrophic lateral sclerosis of Italian origin. J Neurol Neurosurg
Psychiatry. 2007;78(3):327.
105. Praline J, Blasco H, Vourc'h P, et al. Study of the HFE gene common polymorphisms in
French patients with sporadic amyotrophic lateral sclerosis. J Neurol Sci. 2012;317(1-
2):58-61.
106. Lee SY, Patton SM, Henderson RJ, Connor JR. Consequences of expressing mutants of
the hemochromatosis gene (HFE) into a human neuronal cell line lacking endogenous
HFE. The FASEB Journal. 2007;21(2):564-576.
107. Mitchell RM, Lee SY, Simmons Z, Connor JR. HFE polymorphisms affect cellular
glutamate regulation. Neurobiology of Aging. 2011;32(6):1114-1123.
108. Liu Y, Lee SY, Neely E, et al. Mutant HFE H63D protein is associated with prolonged
endoplasmic reticulum stress and increased neuronal vulnerability. The Journal of
biological chemistry. 2011;286(15):13161-13170.
30 109. Nandar W, Neely EB, Simmons Z, Connor JR. H63D HFE genotype accelerates disease
progression in animal models of amyotrophic lateral sclerosis. Biochimica et Biophysica
Acta (BBA) - Molecular Basis of Disease. 2014;1842(12, Part A):2413-2426.
110. Rathore KI, Redensek A, David S. Iron homeostasis in astrocytes and microglia is
differentially regulated by TNF-α and TGF-β1. Glia. 2012;60(5):738-750.
111. Gerber YN, Sabourin JC, Rabano M, Vivanco M, Perrin FE. Early functional deficit and
microglial disturbances in a mouse model of amyotrophic lateral sclerosis. PLoS One.
2012;7(4):e36000.
112. Wang WY, Tan MS, Yu JT, Tan L. Role of pro-inflammatory cytokines released from
microglia in Alzheimer's disease. Ann Transl Med. 2015;3(10):136.
113. Keller JN, Hanni KB, Gabbita SP, Friebe V, Mattson MP, Kindy MS. Oxidized
lipoproteins increase reactive oxygen species formation in microglia and astrocyte cell
lines. Brain Res. 1999;830(1):10-15.
114. Akbas N, Hochstrasser H, Deplazes J, et al. Screening for mutations of the HFE gene in
Parkinson's disease patients with hyperechogenicity of the substantia nigra. Neuroscience
Letters. 2006;407(1):16-19.
115. Aamodt AH, Stovner LJ, Thorstensen K, Lydersen S, White LR, Aasly JO. Prevalence of
haemochromatosis gene mutations in Parkinson’s disease. Journal of Neurology,
Neurosurgery & Psychiatry. 2007;78(3):315-317.
116. Halling J, Petersen MS, Grandjean P, Weihe P, Brosen K. Genetic predisposition to
Parkinson's disease: CYP2D6 and HFE in the Faroe Islands. Pharmacogenetics and
genomics. 2008;18(3):209-212.
117. Golts N, Snyder H, Frasier M, Theisler C, Choi P, Wolozin B. Magnesium Inhibits
Spontaneous and Iron-induced Aggregation of α-Synuclein. Journal of Biological
Chemistry. 2002;277(18):16116-16123.
31 118. Castellani RJ, Siedlak SL, Perry G, Smith MA. Sequestration of iron by Lewy bodies in
Parkinson’s disease. Acta Neuropathol. 2000;100(2):111-114.
119. Roodveldt C, Labrador-Garrido A, Gonzalez-Rey E, et al. Preconditioning of Microglia
by α-Synuclein Strongly Affects the Response Induced by Toll-like Receptor (TLR)
Stimulation. PLoS ONE. 2013;8(11):e79160.
120. Gao HM, Kotzbauer PT, Uryu K, Leight S, Trojanowski JQ, Lee VM.
Neuroinflammation and oxidation/nitration of alpha-synuclein linked to dopaminergic
neurodegeneration. The Journal of neuroscience : the official journal of the Society for
Neuroscience. 2008;28(30):7687-7698.
121. Connor JR, Lee SY. Iron and Cancer. In: Milner J, Donato F, eds. Bioactive Compounds
and Cancer. New York: Humana Press; 2010:469-496.
122. Hann HW, Stahlhut MW, Menduke H. Iron enhances tumor growth. Observation on
spontaneous mammary tumors in mice. Cancer. 1991;68(11):2407-2410.
123. Luanpitpong S, Talbott SJ, Rojanasakul Y, et al. Regulation of lung cancer cell migration
and invasion by reactive oxygen species and caveolin-1. The Journal of biological
chemistry. 2010;285(50):38832-38840.
124. Stevens RG, Morris JE, Cordis GA, Anderson LE, Rosenberg DW, Sasser LB. Oxidative
damage in colon and mammary tissue of the HFE-knockout mouse. Free radical biology
& medicine. 2003;34(9):1212-1216.
125. de Almeida SF, Picarote G, Fleming JV, Carmo-Fonseca M, Azevedo JE, de Sousa M.
Chemical chaperones reduce endoplasmic reticulum stress and prevent mutant HFE
aggregate formation. The Journal of biological chemistry. 2007;282(38):27905-27912.
126. Gunel-Ozcan A, Alyılmaz-Bekmez S, Guler EN, Guc D. HFE H63D mutation frequency
shows an increase in Turkish women with breast cancer. BMC Cancer. 2006;6:37-37.
32 127. Gharib AF, Karam RA, Pasha HF, Radwan MI, Elsawy WH. Polymorphisms of
hemochromatosis, and alpha-1 antitrypsin genes in Egyptian HCV patients with and
without hepatocellular carcinoma. Gene. 2011;489(2):98-102.
128. Elmberg M, Hultcrantz R, Ekbom A, et al. Cancer risk in patients with hereditary
hemochromatosis and in their first-degree relatives. Gastroenterology.125(6):1733-1741.
129. Martinez di Montemuros F, Tavazzi D, Salsano E, et al. High frequency of the H63D
mutation of the hemochromatosis gene (HFE) in malignant gliomas. Neurology.
2001;57(7):1342-1342.
130. Dorak MT, Mackay RK, Relton CL, Worwood M, Parker L, Hall AG. Hereditary
hemochromatosis gene (HFE) variants are associated with birth weight and childhood
leukemia risk. Pediatric Blood & Cancer. 2009;53(7):1242-1248.
131. Shi Z, Johnstone D, Talseth-Palmer BA, et al. Haemochromatosis HFE gene
polymorphisms as potential modifiers of hereditary nonpolyposis colorectal cancer risk
and onset age. International Journal of Cancer. 2009;125(1):78-83.
132. Shaheen NJ, Silverman LM, Keku T, et al. Association Between Hemochromatosis
(HFE) Gene Mutation Carrier Status and the Risk of Colon Cancer. Journal of the
National Cancer Institute. 2003;95(2):154-159.
133. Agudo A, Bonet C, Sala N, et al. Hemochromatosis (HFE) gene mutations and risk of
gastric cancer in the European Prospective Investigation into Cancer and Nutrition
(EPIC) study. Carcinogenesis. 2013;34(6):1244-1250.
134. Ropero P, Briceno O, Lopez-Alonso G, et al. [The H63D mutation in the HFE gene is
related to the risk of hepatocellular carcinoma]. Revista espanola de enfermedades
digestivas : organo oficial de la Sociedad Espanola de Patologia Digestiva.
2007;99(7):376-381.
33 135. Kallianpur AR, Hall LD, Yadav M, et al. Increased prevalence of the HFE C282Y
hemochromatosis allele in women with breast cancer. Cancer epidemiology, biomarkers
& prevention : a publication of the American Association for Cancer Research,
cosponsored by the American Society of Preventive Oncology. 2004;13(2):205-212.
136. Chen W, Zhao H, Li T, Yao H. HFE gene C282Y variant is associated with colorectal
cancer in Caucasians: a meta-analysis. Tumour biology : the journal of the International
Society for Oncodevelopmental Biology and Medicine. 2013;34(4):2255-2259.
137. Lauret E, Rodriguez M, Gonzalez S, et al. HFE gene mutations in alcoholic and virus-
related cirrhotic patients with hepatocellular carcinoma. The American journal of
gastroenterology. 2002;97(4):1016-1021.
138. Fargion S, Stazi MA, Fracanzani AL, et al. Mutations in the HFE gene and their
interaction with exogenous risk factors in hepatocellular carcinoma. Blood cells,
molecules & diseases. 2001;27(2):505-511.
139. Asberg A, Thorstensen K, Irgens WO, Romundstad PR, Hveem K. Cancer risk in HFE
C282Y homozygotes: results from the HUNT 2 study. Scandinavian journal of
gastroenterology. 2013;48(2):189-195.
140. Osborne NJ, Gurrin LC, Allen KJ, et al. HFE C282Y homozygotes are at increased risk
of breast and colorectal cancer. Hepatology. 2010;51(4):1311-1318.
141. Nahon P, Sutton A, Rufat P, et al. Liver iron, HFE gene mutations, and hepatocellular
carcinoma occurrence in patients with cirrhosis. Gastroenterology. 2008;134(1):102-110.
142. Willis G, Wimperis JZ, Lonsdale R, et al. Incidence of liver disease in people with HFE
mutations. Gut. 2000;46(3):401-404.
143. Gannon PO, Medelci S, Le Page C, et al. Impact of hemochromatosis gene (HFE)
mutations on epithelial ovarian cancer risk and prognosis. International Journal of
Cancer. 2011;128(10):2326-2334.
34 144. Rodriguez-Lopez R, Donoso M, Fernandez-Cavada M, et al. Diagnostic utility of HFE
variants in Spanish patients: association with HLA alleles and role in susceptibility to
acute lymphoblastic leukemia. Gene. 2013;514(1):31-35.
145. Abraham BK, Justenhoven C, Pesch B, et al. Investigation of genetic variants of genes of
the hemochromatosis pathway and their role in breast cancer. Cancer epidemiology,
biomarkers & prevention : a publication of the American Association for Cancer
Research, cosponsored by the American Society of Preventive Oncology.
2005;14(5):1102-1107.
146. Altes A, Gimferrer E, Capella G, Barcelo MJ, Baiget M. Colorectal cancer and HFE gene
mutations. Haematologica. 1999;84(5):479-480.
147. Boige V, Castéra L, de Roux N, et al. Lack of association between HFE gene mutations
and hepatocellular carcinoma in patients with cirrhosis. Gut. 2003;52(8):1178-1181.
148. Guruvayoorappan C. Tumor Versus Tumor-Associated Macrophages: How Hot is the
Link? Integrative Cancer Therapies. 2008;7(2):90-95.
149. Kzhyshkowska J, Riabov V, Gudima A, Wang N, Orekhov A, Mickley A. Role of tumor
associated macrophages in tumor angiogenesis and lymphangiogenesis. Frontiers in
Physiology. 2014;5(75).
150. Weston C, Hund W, Nixon, A, et al. Host H67D Genotype Affects Tumor Growth in
Mouse Melanoma. Journal of Cancer Science & Therapy. 2015;7:216-223.
151. Ong SM, Tan YC, Beretta O, et al. Macrophages in human colorectal cancer are pro-
inflammatory and prime T cells towards an anti-tumour type-1 inflammatory response.
European journal of immunology. 2012;42(1):89-100.
152. Muldoon LL, Sandor M, Pinkston KE, Neuwelt EA. Imaging, distribution, and toxicity of
superparamagnetic iron oxide magnetic resonance nanoparticles in the rat brain and
intracerebral tumor. Neurosurgery. 2005;57(4):785-796; discussion 785-796.
35 153. Hashimoto K, Hirai M, Kurosawa Y. Identification of a Mouse Homolog for the Human
Hereditary Haemochromatosis Candidate Gene. Biochemical and Biophysical Research
Communications. 1997;230(1):35-39.
154. Zhou XY, Tomatsu S, Fleming RE, et al. HFE gene knockout produces mouse model of
hereditary hemochromatosis. Proceedings of the National Academy of Sciences.
1998;95(5):2492-2497.
155. Bahram S, Gilfillan S, Kühn LC, et al. Experimental hemochromatosis due to MHC class
I HFE deficiency: Immune status and iron metabolism. Proceedings of the National
Academy of Sciences of the United States of America. 1999;96(23):13312-13317.
156. Levy JE, Montross LK, Cohen DE, Fleming MD, Andrews NC. The C282Y Mutation
Causing Hereditary Hemochromatosis Does Not Produce a Null Allele. Blood.
1999;94(1):9-11.
157. Tomatsu S, Orii KO, Fleming RE, et al. Contribution of the H63D mutation in HFE to
murine hereditary hemochromatosis. Proceedings of the National Academy of Sciences.
2003;100(26):15788-15793.
158. Merryweather-Clarke AT, Pointon JJ, Shearman JD, Robson KJ. Global prevalence of
putative haemochromatosis mutations. Journal of medical genetics. 1997;34(4):275-278.
159. Pulliam JF, Jennings CD, Kryscio RJ, et al. Association of HFE mutations with
neurodegeneration and oxidative stress in Alzheimer's disease and correlation with
APOE. American journal of medical genetics Part B, Neuropsychiatric genetics : the
official publication of the International Society of Psychiatric Genetics. 2003;119b(1):48-
53.
160. Warner RC, Weber I. The Cupric and Ferric Citrate Complexes1. Journal of the
American Chemical Society. 1953;75(20):5086-5094.
36 161. Brittenham G. The red cell cycle. In: Brock J, ed. Iron Metabolism in Health and
Disease. London: W.B. Saunders Company Ltd; 1994:31–62.
162. Soe-Lin S, Apte SS, Mikhael MR, Kayembe LK, Nie G, Ponka P. Both Nramp1 and
DMT1 are necessary for efficient macrophage iron recycling. Experimental Hematology.
2010;38(8):609-617.
163. McKie AT, Marciani P, Rolfs A, et al. A novel duodenal iron-regulated transporter,
IREG1, implicated in the basolateral transfer of iron to the circulation. Molecular cell.
2000;5(2):299-309.
164. Connor JR, Zhang X, Nixon AM, Webb B, Perno JR. Comparative evaluation of
nephrotoxicity and management by macrophages of intravenous pharmaceutical iron
formulations. PLoS One. 2015;10(5):e0125272.
165. Zhang X, Surguladze N, Slagle-Webb B, Cozzi A, Connor JR. Cellular iron status
influences the functional relationship between microglia and oligodendrocytes. Glia.
2006;54(8):795-804.
166. Cohen LA, Gutierrez L, Weiss A, et al. Serum ferritin is derived primarily from
macrophages through a nonclassical secretory pathway. Blood. 2010;116(9):1574-1584.
167. Zhang X, Haaf M, Todorich B, et al. Cytokine toxicity to oligodendrocyte precursors is
mediated by iron. Glia. 2005;52(3):199-208.
168. Goodman GT, Koprowski H. Macrophages as a cellular expression of inherited natural
resistance. Proceedings of the National Academy of Sciences of the United States of
America. 1962;48(2):160-165.
169. Mackaness GB. Cellular resistance to infection. The Journal of experimental medicine.
1962;116:381-406.
37 170. Gratchev A, Guillot P, Hakiy N, et al. Alternatively activated macrophages differentially
express fibronectin and its splice variants and the extracellular matrix protein betaIG-H3.
Scandinavian journal of immunology. 2001;53(4):386-392.
171. Mackaness GB. Cellular immunity and the parasite. Advances in experimental medicine
and biology. 1977;93:65-73.
172. Nathan CF, Murray HW, Wiebe ME, Rubin BY. Identification of interferon-gamma as
the lymphokine that activates human macrophage oxidative metabolism and
antimicrobial activity. The Journal of experimental medicine. 1983;158(3):670-689.
173. Arango Duque G, Descoteaux A. Macrophage Cytokines: Involvement in Immunity and
Infectious Diseases. Frontiers in Immunology. 2014;5:491.
174. Corna G, Campana L, Pignatti E, et al. Polarization dictates iron handling by
inflammatory and alternatively activated macrophages. Haematologica.
2010;95(11):1814-1822.
175. Wrighting DM, Andrews NC. Interleukin-6 induces hepcidin expression through STAT3.
Blood. 2006;108(9):3204-3209.
176. Weinberg ED. Iron withholding: a defense against infection and neoplasia. Physiological
Reviews. 1984;64(1):65-102.
177. Edwards JP, Zhang X, Frauwirth KA, Mosser DM. Biochemical and functional
characterization of three activated macrophage populations. Journal of Leukocyte
Biology. 2006;80(6):1298-1307.
178. Albina JE, Mills CD, Henry WL, Jr., Caldwell MD. Temporal expression of different
pathways of 1-arginine metabolism in healing wounds. Journal of immunology
(Baltimore, Md : 1950). 1990;144(10):3877-3880.
38 179. Nagaoka T, Kaburagi Y, Hamaguchi Y, et al. Delayed wound healing in the absence of
intercellular adhesion molecule-1 or L-selectin expression. The American journal of
pathology. 2000;157(1):237-247.
180. Subramaniam M, Saffaripour S, Van De Water L, et al. Role of endothelial selectins in
wound repair. The American journal of pathology. 1997;150(5):1701-1709.
181. Martin P, D'Souza D, Martin J, et al. Wound Healing in the PU.1 Null Mouse—Tissue
Repair Is Not Dependent on Inflammatory Cells. Current Biology. 2003;13(13):1122-
1128.
182. Sica A, Schioppa T, Mantovani A, Allavena P. Tumour-associated macrophages are a
distinct M2 polarised population promoting tumour progression: Potential targets of anti-
cancer therapy. European Journal of Cancer. 2006;42(6):717-727.
183. Valberg LS, Simon JB, Manley PN, Corbett WE, Ludwig J. Distribution of storage iron
as body iron stores expand in patients with hemochromatosis. Translational
Research.86(3):479-489.
184. Brink B, Disler P, Lynch S, Jacobs P, Charlton R, Bothwell T. Patterns of iron storage in
dietary iron overload and idiopathic hemochromatosis. The Journal of laboratory and
clinical medicine. 1976;88(5):725-731.
185. Montosi G, Paglia P, Garuti C, et al. Wild-type HFE protein normalizes transferrin iron
accumulation in macrophages from subjects with hereditary hemochromatosis. Blood.
2000;96(3):1125-1129.
186. Drakesmith H, Sweetland E, Schimanski L, et al. The hemochromatosis protein HFE
inhibits iron export from macrophages. Proceedings of the National Academy of
Sciences. 2002;99(24):15602-15607.
39 187. Townsend A, Drakesmith H. Role of HFE in iron metabolism, hereditary
haemochromatosis, anaemia of chronic disease, and secondary iron overload. Lancet.
2002;359(9308):786-790.
2
Chapter 2
Research Gaps
2.1 Research Gaps
The discovery of the HFE H63D mutation was due to increased parenchymal iron found within patients1. In addition, macrophages, which normally store iron, are iron poor when HFE is mutated2. The reason behind iron poor macrophages and increased parenchymal iron accumulation is still unknown. It has been hypothesized that decreased macrophage iron within
HFE mutants is due to lower transferrin receptor levels or disruption of the iron export mechanism3-5. However, these previous results either looked only at wildtype HFE or used a transfected monocytic cell line, which may not accurately portray the physiological role of an
HFE mutation. Therefore, our first research chapter explores how H63D mutation effects iron handling in a primary macrophage cell culture line. Figure 2-1 depicts the hypothesized results of the H67D HFE mutation, mouse homolog of the human H63D mutation, on bone marrow macrophages phenotype.
41 Figure 2-1
H67D Macrophage WT Macrophage
Figure 2-1: Hypothesized Chapter 3 Results. Hypothesized results suggest that the H67D mutation within macrophages would result in increased transferrin receptor and ferroportin, and decreased levels of ferritin corresponding to the increased parenchymal iron, and decreased macrophage iron physiologically reported in the human population.
42 The H63D HFE mutation extends greater than the physiological disruption of iron homeostasis.
Previous research has reported the human H63D and mouse H67D mutations result in increased oxidative stress, brain iron, and neuronal vulnerability6,7. The results of these studies, led to the question of how H63D mutations effect disease progression. Literature has reported an association of H63D mutations and Alzheimer’s Disease and Amyotrophic Lateral Sclerosis8-13, however there have not been any studies looking at the relationship of H63D/H67D mutations and Parkinson’s Disease. Due to the increased stress reported in H67D mouse models, we hypothesize that in a Parkinsonism model induced by paraquat, the H67D mice would have exacerbated effects as predicted in Figure 2-2. In addition, the association between cholesterol, statins, and Parkinson’s Disease has been of increased interest. There has been conflicting literature whether cholesterol increases the risk for Parkinson’s Disease14, whereas simvastatin is thought to be protective against15. Furthermore, this relationship may be influenced by the H67D mutation as previous reports have found H67D mutated mice have altered cholesterol levels and accelerated ALS disease progression with simvastatin16,17. Therefore, Chapter 5 addresses how a cholesterol or simvastatin affects motor function in a H67D parkinsonism paraquat induced model. The hypothesized results are reported in Figure 2-3.
43 Figure 2-2 Paraquat i.p. Saline i.p.
H67D WT H67D WT • ↑ ↑ Microglia • ↑ Microglia • ↑ Microglia • ⟷ Microglia • ↓ ↓ Motor behavior • ↓ Motor behavior • ⟷ Motor behavior • ⟷ Motor behavior • ↓ ↓ TH neurons • ↓ TH neurons • ⟷ TH neurons • ⟷ TH neurons • ↑ ↑ Iron • ↑ Iron • ↑ Iron • ⟷ Iron
Figure 2-2: Hypothesized Chapter 4 Results. Upon injection of paraquat, it is hypothesized that the H67D mice will have increased neurological consequences, such as decreased motor function due to increased tyrosine hydroxylase (TH) neuron loss. The loss of TH neurons should be accompanied by an increase in microglia and iron. The WT mice treated with paraquat will have decreased neurological function as well, however not as significant as the H67D mice.
44 Figure 2-3 A Paraquat i.p. Saline i.p.
Control Diet Control Diet
H67D WT H67D WT • ↓ ↓ Motor behavior • ↓ Motor behavior • ⟷ Motor behavior • ⟷ Motor behavior
B Paraquat i.p. Saline i.p.
Cholesterol Diet Cholesterol Diet
H67D WT H67D WT • ↓ ↓ ↓ Motor behavior • ↓ ↓ Motor behavior • ⟷ Motor behavior • ⟷ Motor behavior
C Paraquat i.p. Saline i.p.
Simvastatin Diet Simvastatin Diet
H67D WT H67D WT • ↓ Motor behavior • ⟷ Motor behavior • ⟷ Motor behavior • ⟷ Motor behavior
45 Figure 2-3: Hypothesized Chapter 5 Results. WT mice treated with paraquat on a control diet, are hypothesized to have decreased motor behavior, however the counterpart H67D mice will have further decreased motor behavior (A). When on a cholesterol diet, the mice will have exaggerated behavioral effects (B), whereas on the simvastatin diet that mice will exhibit a neuroprotective effect against the paraquat treatment (C).
46 2.2 References
1. Simon M, Pawlotsky Y, Bourel M, Fauchet R, Genetet B. Hemochromatose idiopathique:
maladie associee a l'anligine tissulaire HLA 3? La Nouvelle presse medicale.
1975;4(19):1432.
2. Valberg LS, Simon JB, Manley PN, Corbett WE, Ludwig J. Distribution of storage iron
as body iron stores expand in patients with hemochromatosis. Translational
Research.86(3):479-489.
3. Brink B, Disler P, Lynch S, Jacobs P, Charlton R, Bothwell T. Patterns of iron storage in
dietary iron overload and idiopathic hemochromatosis. The Journal of laboratory and
clinical medicine. 1976;88(5):725-731.
4. Montosi G, Paglia P, Garuti C, et al. Wild-type HFE protein normalizes transferrin iron
accumulation in macrophages from subjects with hereditary hemochromatosis. Blood.
2000;96(3):1125-1129.
5. Drakesmith H, Sweetland E, Schimanski L, et al. The hemochromatosis protein HFE
inhibits iron export from macrophages. Proceedings of the National Academy of
Sciences. 2002;99(24):15602-15607.
6. Nandar W, Neely EB, Unger E, Connor JR. A mutation in the HFE gene is associated
with altered brain iron profiles and increased oxidative stress in mice. Biochimica et
Biophysica Acta (BBA) - Molecular Basis of Disease. 2013;1832(6):729-741.
7. Liu Y, Lee SY, Neely E, et al. Mutant HFE H63D protein is associated with prolonged
endoplasmic reticulum stress and increased neuronal vulnerability. The Journal of
biological chemistry. 2011;286(15):13161-13170.
8. Moalem S, Percy ME, Andrews DF, et al. Are hereditary hemochromatosis mutations
involved in Alzheimer disease? American journal of medical genetics. 2000;93(1):58-66.
47 9. Pulliam JF, Jennings CD, Kryscio RJ, et al. Association of HFE mutations with
neurodegeneration and oxidative stress in Alzheimer's disease and correlation with
APOE. American journal of medical genetics Part B, Neuropsychiatric genetics : the
official publication of the International Society of Psychiatric Genetics. 2003;119b(1):48-
53.
10. Correia AP, Pinto JP, Dias V, Mascarenhas C, Almeida S, Porto G. CAT53 and HFE
alleles in Alzheimer's disease: a putative protective role of the C282Y HFE mutation.
Neurosci Lett. 2009;457(3):129-132.
11. Blazquez L, De Juan D, Ruiz-Martinez J, et al. Genes related to iron metabolism and
susceptibility to Alzheimer's disease in Basque population. Neurobiol Aging.
2007;28(12):1941-1943.
12. Wang X-S, Lee S, Simmons Z, et al. Increased incidence of the Hfe mutation in
amyotrophic lateral sclerosis and related cellular consequences. Journal of the
Neurological Sciences. 2004;227(1):27-33.
13. Goodall EF, Greenway MJ, van Marion I, Carroll CB, Hardiman O, Morrison KE.
Association of the H63D polymorphism in the hemochromatosis gene with sporadic
ALS. Neurology. 2005;65(6):934-937.
14. Bosco DA, Fowler DM, Zhang Q, et al. Elevated levels of oxidized cholesterol
metabolites in Lewy body disease brains accelerate alpha-synuclein fibrilization. Nature
chemical biology. 2006;2(5):249-253.
15. Bar-On P, Crews L, Koob AO, et al. Statins reduce neuronal α-synuclein aggregation in
in vitro models of Parkinson's disease. Journal of neurochemistry. 2008;105(5):1656.
16. Ali-Rahmani F, Grigson PS, Lee S, Neely E, Connor JR, Schengrund CL. H63D
mutation in hemochromatosis alters cholesterol metabolism and induces memory
impairment. Neurobiol Aging. 2014;35(6):1511.e1511-1512.
48 17. Su XW, Nandar W, Neely EB, Simmons Z, Connor JR. Statins accelerate disease
progression and shorten survival in SOD1(G93A) mice. Muscle & nerve.
2016;54(2):284-291.
49 Chapter 3
The Role of HFE Genotype in Macrophage Phenotype
3.1 Abstract
As an essential component for cellular energy production and general homeostasis, iron regulation has critical implications in both normal human function and disease. The HFE mutation, the most common gene variant in Caucasians, alters cellular iron homeostasis and is frequently associated with a number of neurological diseases and cancer. One cell type whose function is critical to both maintaining iron homeostasis and the pathogenic process in multiple diseases is the macrophage. This study addresses whether macrophage HFE genotype affects macrophage function. The results of this study demonstrate that mutations in the HFE genotype significantly impact a number of critical macrophage functions, but differences in iron handling are minimal in the “resting” macrophages. Specifically, proliferation in response to iron exposure,
L-ferritin expression in response to iron, secretion of BMP6 and cytokines were all functions found to be genotype specific. In addition to the HFE mutant specific changes to cytokine secretion profiles, two significant components of the immune response, migration and phagocytosis, were affected by changes in genotype as well. A significant observation, although not genotype specific, was the demonstration that exposure to apo-Tf (iron-poor transferrin) can increase the release of iron from macrophages. Because under normal conditions, 70% of circulating transferrin in unsaturated, the ability of apo-Tf to induce iron release could be a major regulatory mechanism for iron release from macrophages. These studies identify areas for further studies into HFE impact via macrophage dysfunction in the context of neurodegenerative disorders, in addition to all types of cancer.
50 3.2 Introduction
Iron is critical cofactor in many biological and physiological processes1. Therefore, iron mismanagement can lead to dysfunction in multiple systems and damage. In the past twenty years, significant attention has been focused upon mutations in the HFE (high iron) gene that can lead to increased total body iron and thus detrimental effects within the body.
The HFE gene was discovered to encode the HFE protein, a transmembrane glycoprotein, similar to MHC molecules2. HFE is thought to contribute to the regulation of iron through its ability to bind to the transferrin receptor on the cellular membrane3,4. In support of this, mutations within this gene result in systemic aberrant iron storage termed, hemochromatosis. There are four different types of hemochromatosis mutations, each with distinct underlying molecular mechanisms. Importantly, type I hemochromatosis is the only disorder caused by a mutation within the HFE gene2. However, the mechanisms in which mutated HFE causes type 1 hemochromatosis remains undefined. Hence, further elucidation and characterization of functional role the HFE mutations may allow for novel therapies toward type 1 hemochromatosis.
There are three known polymorphisms of the HFE gene; H63D, C282Y, and S56C of which
C282Y is the most prevalent in those patients with hemochromatosis, however H63D is the most common mutation, especially in the Caucasian population2,5-7.
One of the key characteristics of hereditary hemochromatosis is that reticuloendothelial cells, specifically macrophages, are reportedly iron poor8,9. This is of significance as macrophages play an important role in iron homeostasis through recycling iron, in addition to maintaining immune defense10-13. Neither the reason for nor the mechanism underlying the atypical iron levels within
HFE mutated macrophages is known. It has been suggested that the HFE protein interacts differently with parenchymal cells compared to macrophages, in that mutated HFE can prevent
51 iron uptake14,15 in the macrophages. Moreover, it is speculated that HFE functions to regulate hepcidin levels through bone morphogenetic protein 6 pathway (BMP6), which would result in the stimulation of hepcidin16. Thus, loss of HFE function may reduce hepcidin production which would result in greater iron uptake from the gut17. Clearly, the precise elucidation of the role of
HFE mutations in macrophages is important.
Furthermore, the functional implications of mutated HFE protein expands beyond hemochromatosis, as the mutation is reportedly increased in neurodegenerative diseases such as
Alzheimer’s Disease and ALS18-23. Importantly, microglia, the resident macrophage of the brain, has been shown to be activated in all neurodegenerative diseases 24. In addition, HFE mutations are also linked to increased frequency of cancer25,26and macrophages densely populate tumors
27,28. Thus iron regulation and its relationship to macrophage function may provide an underlying mechanistic link between otherwise unrelated diseases that are influenced by HFE genotype. For example, without iron sequestration in the macrophages, free radicals and oxidative stress could increase in primary cells of the brain or cancer cells, promoting disease. In this paper, we address how HFE mutations influence the phenotype of macrophages.
3.3 Materials and Methods
3.3.1 Mouse Colony
C57BL/6J x 129 mice (12-month old males) that contained either wildtype (WT) or H67D, a homolog for the human H63D HFE mutation, alleles were used as previously described29. The mice were maintained in an in-house animal facility at The Pennsylvania State University College of Medicine, and all procedures were approved by the Pennsylvania State University College of
Medicine Institutional Animal Care and Use Committee, protocol 04-166.
52 3.3.2 Primary Macrophage Culture
Bone marrow-derived cells were extracted and cultured as previously described30. Briefly, the mice were sacrificed by cervical dislocation. Next, the femurs and tibias were removed and their epiphyses excised. The bone marrow was then flushed using Dulbecco’s phosphate-buffered saline (DPBS) (Corning; Manassas, VA). The isolated cell suspension was passed through a 40
�m cell strainer. The bone marrow cells were then plated in 100 mm2 tissue culture plates at a concentration of 8x106 cells/plate. The cells were maintained in Dulbecco's Modified Eagle's
Medium (DMEM) (Invitrogen; Grand Island, NY) with 10% fetal bovine serum (FBS) (Gemini
Bio Products; West Sacramento, CA), 1% penicillin-streptomycin (Invitrogen; Grand Island,
NY), and 10 ng/ml macrophage colony stimulating factory (M-CSF) (R&D Systems;
Minneapolis, MN), and incubated at 37°C under an atmosphere of 5% CO2 with humidified air.
M-CSF was supplemented to the cell culture media to allow differentiation of the bone marrow cells to bone marrow-derived macrophages (BMMs). BMMs were cultured for 10 days until approximately 100% of cells have derived into macrophages. Verification was confirmed through flow cytometry. Following, the BMMs were used for various experiments explained below.
3.3.3 MTT and LDH Assays
To determine cell viability of the BMMs, 3-(4,5-Dimethylthiazol-2-Yl)-2,5-Diphenyltetrazolium
Bromide (MTT) and lactate dehydrogenase (LDH) were used. Following culturing of BMMs, the cells were washed with DPBS and then divided into three different groups that received either control media (same as culture media), media supplemented with 300 uM ferric ammonium citrate (FAC) (Sigma Aldrich; St. Louis, MO) for iron loading, or 300 uM deferoxamine (DFO)
(Sigma Aldrich; St. Louis, MO), an iron chelator. Following a 24-hour incubation, the cells were washed with DPBS. The media of all of the cell culture groups were then replaced with control media and incubated for an additional 24-hours. Subsequently, cell viability was assessed using
53 MTT (Roche; Basal, Switzerland) and LDH (Roche; Basal, Switzerland) assays according to manufactures instructions. Fluorescence of the cells was measured on a SpectraMax Gemini EM plate reader (Molecular Devices; Sunnyvale, CA) at the corresponding wave length.
3.3.4 Enzyme Linked Immunosorbent Assays (ELISAs)
After differentiation, BMMs were divided into three treatment groups receiving control, 300 uM
FAC, or 300 uM DFO supplemented media. The BMMs were incubated for 24 hours and then washed with DPBS. The treatment media was then replaced with control media, which all groups received. The BMMs were incubated for another 24 hours. Afterwards the media was collected and the cells were washed twice with DPBS. The macrophages were then lysed in RIPA buffer
(Sigma Aldrich; St. Louis, MO) with a 1:100 dilution of protease inhibitors (Sigma Aldrich; St.
Louis, MO). Protein concentrations were determined using a BCA protein assay kit (Pierce;
Rockford, IL).
3.3.4.1 Iron Proteins
The cell lysates were used to measure the amount of iron proteins present in wildtype and H67D macrophages. ELISAs for the iron proteins included L-ferritin (Abcam; Cambridge, United
Kingdom), H-ferritin (Mybiosource; San Diego, CA), transferrin receptor (BlueGene; Shanghai,
China), and ferroportin (Cloud-Clone Corp; Houston, TX) and performed according to their manufacturer’s instructions. Absorbance of the cell lysates was measured on a SpectraMax
340PC plate reader (Molecular Devices; Sunnyvale, CA)
3.3.4.2 Bone Morphogenetic Protein 6 (BMP6) and Bone Morphogenetic Type I Receptor (ALK3)
To quantify the amount of BMP6 and ALK3 was present within the macrophages, and how much was secreted from the macrophages, the cells were lysed and their media was collected. The
54 samples were then run on a BMP6 (DL Develop; Wuxi, Jiangsu, China) and ALK3
(Mybiosource; San Diego, CA) ELISA according to their manufactures instructions. Absorbance of the cell lysates and cell culture media was measured on a SpectraMax 340PC plate reader
(Molecular Devices; Sunnyvale, CA) at the appropriate wave length.
3.3.5 Phagocytosis Assay
A phagocytosis assay (Molecular Probes; Eugene, OR) was used to examine the level of phagocytosis in each macrophage genotype. Briefly, 1x104 BMMs/wells were plated on a black bottom 96-well plate. The cells were either incubated with cell culture media for 24 hours.
Following, the media was replaced with fluorescently labeled E. coli bioparticles for an additional two hours. Afterwards the bioparticles were aspirated and excess fluorescence quenched by an addition of trypan blue. Fluorescence of the cells was measured on a SpectraMax Gemini EM plate reader (Molecular Devices; Sunnyvale, CA) at 480 nm/520 nm.
3.3.6 Comparison of Cellular Migration
Cellular migration of BMMs was assessed using a Cytoselect Cell Invasion Assay (Cell Bio Labs,
Inc.; San Diego, CA), consisting of a 5 um membrane insert. One million BMMs in serum free media were added to the membrane insert, in which the lower well of the migration plate contained FBS supplemented cell culture media. The cell suspension was incubated for 24-hours at 37°C. Following incubation, the cells that had migrated into the lower well were lysed using a fluorescent lysis buffer. Fluorescence of the samples was measured on a SpectraMax Gemini EM plate reader (Molecular Devices; Sunnyvale, CA) at 480 nm/520 nm.
55 3.3.7 Analysis of Macrophage Cytokines
Pro-inflammatory cytokine levels of BMMs lysates were assessed with a mouse cytokine array kit (R&D Systems; Minneapolis, MN). BMMs were incubated with either control media, 50 ng/ml of lipopolysaccharide (LPS) (Sigma Aldrich; St. Louis, MO), or 100 uM paraquat dichloride hydrate (Sigma Aldrich; St. Louis, MO) for 24 hours. Following incubation, the cells were washed with PBS and lysed with the assay lysis buffer. The lysates were then mixed with a cocktail of biotinylated antibodies and then incubated on a nitrocellulose membrane containing 40 different anti pro-inflammatory cytokine antibodies. Following, the membrane was washed and then incubated with streptavidin-HRP for 30 minutes. The membranes were then processed further following the manufactures instructions and exposed using GE Amersham Imager 600
(GE; Buckinghamshire, United Kingdom). The results were then analyzed using ImageJ software
(NIH; Bethesda, MD).
3.3.8 59Fe Loading and Release
The amount of iron released from BMMs was measured by loading the cells with 59Fe and monitoring their release over 24 hours. One million BMMs were plated in a 6-well tissue plate and incubated with 2 uCi/well of 59Fe-NTA complex, overnight. The 59Fe-NTA complex was generated as previously described31. Following, the cells were washed twice with DPBS, to remove iron from the culture media, and replaced with different conditioned treatment media.
The media was replaced with either control, control/500 nM hepcidin, 50 mg/ml apo- transferrin(Apo-Tf), Apo-Tf/hepcidin, 20 uM DFO, or DFO/hepcidin media. Data were collected by taking 100 uL aliquots at 0, 4, 8, 12, and 24 hours. The amount of 59Fe within the samples was then measured on a Beckman Gamma 4000 (Beckman Coulter; Brea, CA).
56 3.3.9 Statistical Analysis
All statistical analyses were analyzed using the Graphpad Prism (La Jolla, CA), in which the results were presented as mean ± SEM. Statistical comparisons were completed by either a ungrouped t-test or a 1-way ANOVA. A p-value of ≤0.05 was considered statistically significant.
3.4 Results
3.4.1 Cellular Proliferation of Wildtype Macrophages Increases in Iron Rich Conditions
To determine the effects of exogenous iron treatments on the growth rate of H67D mutant macrophages, we assessed cell growth and survival of macrophage genotype in the presence of
300 uM FAC, 300 uM DFO, as well as normal culture conditions via MTT and LDH assay respectively. Wildtype BMMs treated with 300 uM FAC had a 24% increase (p<0.01) in cell proliferation over the non-treated and DFO mediated iron-depleted wildtype BMMs, indicating increased iron levels promotes growth in normal BMMs while iron deprivation has minimal effect (Fig. 3-1A). However, FAC mediated iron loading failed to induce proliferation in H67D
BMMs, suggesting the mutation inhibits growth in iron rich environments, compared to WT control BMMs. The LDH assay showed the treatments did not result in significant cellular death
(Fig. 3-1B).
57 Figure 3-1 MTT LDH A B 150 150 **
100 100
50 50 % of Control % of Control
0 0
Wt Control Wt Control H67D Control H67D Control Wt 300 uMWt FAC 300 uM DFO Wt 300 uMWt FAC 300 uM DFO H67D 300 H67DuM FAC 300 uM DFO H67D 300 H67DuM FAC 300 uM DFO
Figure 3-1: Cell Proliferation (MTT) and Cytotoxicity (LDH) of BMMs Incubated with
Control, FAC, or DFO Media. MTT (A) and LDH (B) assays were carried out to determine cell proliferation and cellular toxicity, respectively. The cells were exposed to control, 300 uM FAC, or 300 uM DFO supplemented media for 24 hours. Following the media was replaced with control media for an additional 24 hours, and cell proliferation/cytotoxicity was assessed. Data represent the mean ± SEM from three independent experiments and are compared to all treatment groups for statistical significance using 1-way ANOVA. **: compared to control p<0.01.
58 3.4.2 Iron loading promotes L-ferritin expression in H67D mutant BMMs
To further characterize the effects of iron loading on H67D mutant BMMs, we assessed the expression of several key iron regulatory proteins in iron rich and iron poor conditions.
Specifically, we assessed the expression levels L-ferritin, H-ferritin, transferrin receptor, and ferroportin in the lysate of BBMs cells incubated in the presence of 300 uM FAC or 300 uM
DFO. L-ferritin expression was increased following exposure to iron in the media in both WT and H67D BMMs. However, there was a genotype specific difference with the mutant macrophages increasing by 101% (p<0.0001) compared to WT (Fig. 3-2A). DFO exposure had no effect on L-ferritin expression. Interestingly, the other iron storage protein, H-ferritin, did not show any significant differences between treatment groups and genotype following iron exposure or exposure to the iron chelator (Fig. 3-2B). Ferroportin was detected in resting macrophages at the same level regardless of genotype. In response to iron treatment, both wildtype and mutant macrophages had similar increases in expression of ferroportin. Exposure to the iron chelator,
DFO, had no effect on expression of ferroportin (Fig. 3-2C). Lastly, assessment of the transferrin receptor yielded high levels of transferrin receptor expression across all groups. However, there were no trends or significant differences between genotypes or treatment groups (Fig. 3-2D).
59 Figure 3-2
20 25 A **** B H67D Control H67D Control 20 H67D 300 uM FAC H67D 300 uM FAC 15 H67D 300 uM DFO H67D 300 uM DFO 15 Wt Control Wt Control 10 Wt 300 uM FAC Wt 300 uM FAC 10 Wt 300 uM DFO Wt 300 uM DFO 5 5 Intracellular H-ferritin (ng) H-ferritin Intracellular Intracellular L-ferritin (ug ) (ug L-ferritin Intracellular 0 0
Wt Control Wt Control H67D Control H67D Control Wt 300 uMWt FAC 300 uM DFO Wt 300 uMWt FAC 300 uM DFO H67D 300 H67DuM FAC 300 uM DFO H67D 300 H67DuM FAC 300 uM DFO
C 100 D 10 H67D Control 80 H67D 300 uM FAC H67D Control 8 H67D 300 uM DFO H67D 300 uM FAC 60 Wt Control H67D 300 uM DFO 6 Wt 300 uM FAC Wt Control 40 Wt 300 uM FAC 4 Wt 300 uM DFO 20 Wt 300 uM DFO 2
Intracellular Ferroportin (ng) Ferroportin Intracellular 0 0
Wt Control (ng) Receptor Transferrin Intracellular H67D Control Wt 300 uMWt FAC 300 uM DFO Wt Control H67D 300 H67DuM FAC 300 uM DFO H67D Control Wt 300 uMWt FAC 300 uM DFO H67D 300 H67DuM FAC 300 uM DFO
Figure 3-2: Iron Handling Protein Concentrations in Cellular Lysates. BMMS were plated in
100 mm2 tissue culture dishes (8 x 106 cells/flask). Following maturation, the cells were incubated with control, 300 uM FAC, or 300 uM DFO supplemented media for 24 hours. Following the media was replaced with control media for an additional 24 hours. After 24 hours, cells were harvested and lysed for analysis using an L-ferritin (a), H-ferritin (b), ferroportin (c), transferrin- receptor (d) ELISA. Absorbance of the cell lysates was measured on a SpectraMax 340PC plate
60 reader. Data represent the mean±SEM from three independent experiments and are compared to all treatment groups for statistical significance using 1-way ANOVA. ****p<0.0001
61 3.4.3 H67D mutant promotes BMP6 secretion
The expression of BMP6 and its receptor ALK3, two important proteins potentially involved in the regulation of HFE16, were assessed in H67D and WT BMMs. BMP6 was measured in both the cell culture media and cell lysates. There was no genotype difference in intracellular BMP6
(Fig. 3-3A). However, there were significant differences found in the cell culture media, in which the H67D macrophages secreted increased levels of BMP6 compared to wildtype macrophages
(Fig. 3-3B). In addition, there was no genotype difference between wildtype and mutant in intracellular expression of ALK3. (Fig. 3-3C).
62 Figure 3-3
A BMP6 B BMP6 Conditioned Media 1000 800 H67D 800 WT * 600 600 400 pg/ml 400 pg/ml
200 200
0 0
WT H67D H67D Wildtype C AIK3 0.25 H67D 0.20 WT
0.15
ng/ml 0.10
0.05
0.00
WT H67D
Figure 3-3: BMP6 and BMP6 Receptor, ALK3, Concentration in Cellular Culture Media and Cellular Lysates. BMMS were plated in 100 mm2 tissue culture dishes (8 x 106 cells/dish).
Following maturation, the cell culture media was replaced with fresh control media for 24 hours.
After 24 hours, the media was collected and cells were harvested and lysed for analysis. The cell culture media and cell lysates were used for a BMP6 or ALK3 ELISA. Absorbance of the cell lysates was measured on a SpectraMax 340PC plate reader. Data represent the mean ± SEM from three independent experiments and statistical significance completed using an unpaired t-test with
Welch’s correction. *p<0.05
63 3.4.4 Radioactive Iron Release
Iron release was measured by loading the macrophages with radioactive iron 59Fe, overnight. No significant genotype differences were observed throughout the course of the experiment.
However, there were significant changes between the treatment groups beginning at hour 4 (data not shown). The most significant differences occurred at 24 hours (Fig. 3-4). Overall, macrophages, independent of genotype, released iron when exposed to apo-Tf or the iron chelator
DFO.
64 Figure 3-4
800
**** 600 **** **** ** 400 H67D DFO (CPM) H67D DFO/Hepcidin 200 WT Control Counts Per Minute WT Control/Hepcidin 0 WT Apo-Tf Wt Apo-Tf/Hepcidin
WT DFO WT DFO H67D DFO WT Control WT Apo-Tf H67D Control H67D Apo-Tf WT DFO/Hepcidin
WT DFO/Hepcidin Wt Apo-Tf/Hepcidin H67D DFO/HepcidinWT Control/Hepcidin H67D Control/HepcidinH67D Apo-Tf/Hepcidin
Figure 3-4: Amount of 59Fe released from BMMs following 24 hours of 59Fe loading.
2 uCi/well of 59Fe was loaded to 1 x 106 BMMs, plated in 6-well tissue culture dishes, overnight.
Following the cells were washed and replaced with media containing hepcidin, apo-Tf, DFO, apo-Tf and hepcidin, or DFO and hepcidin. 100 ul aliquots were taken at each time point. Data represent the mean±SEM from three independent experiments and are compared to control for statistical significance using two-way ANOVA. **p<0.01; ****p<0.0001.
65 3.4.5 H67D regulates migration, phagocytosis and inflammatory cytokine expression in BMMs
To characterize the functional significance of H67D in BMMs, we assessed several major functions of macrophages; specifically, migration, phagocytosis, and expression of inflammatory cytokines. To determine if the HFE genotype affects cellular migration, BMMs were used in a chemotaxis assay in which FBS was used as the chemoattractant. There were significantly more
WT macrophages that migrated across the membrane compared to H67D macrophages, indicating the H67D mutation hinders migration in BMM. (Fig. 3-5). The rate of phagocytosis between the two genotypes was assessed, in which H67D macrophages have approximately twice the amount of phagocytosis activity (p<0.0001) compared to the wildtype macrophages (Fig. 3-6).
66 Figure 3-5
Figure 3-5: Cellular Migration of the Chemoattractant FBS. One million BMMS suspended in serum free media were placed in a 4 um membrane separated from the chemoattractant FBS, located in bottom well. The cells that migrated across the membrane were lysed with a fluorescence lysis buffer. Fluorescence was quantified on a fluorescent plate reader at excitation/emission wavelengths of 490/520 nm. Data represent the mean ± SEM from three independent experiments and statistical significance completed using an unpaired t-test with
Welch’s correction. *p<0.05.
67 Figure 3-6
100 **** WT 80 H67D
60
40
nm 480/520 OD 20
0
WT H67D
Figure 3-6: Resting State Phagocytosis Activity
BMM phagocytosis, or cell culture media without treatment. Following a 24-hour incubation, the cells were then incubated with a fluorescently labeled E. coli bioparticle. The fluorescence of the engulfed bioparticles was quantified on a fluorescent plate reader at excitation/emission wavelengths of 480/520 nm. Data represent the mean ± SEM from three independent experiments and statistical significance completed using an unpaired t-test with Welch’s correction.
*p<0.0001.
68 To determine if H67D alters inflammatory cytokine expression, 40 different pro-inflammatory cytokines were analyzed using a cytokine array panel on BMM macrophages treated with control,
50 ng/ml LPS, or 100 uM paraquat. Through this panel, 18 cytokines were detected within the
BMM cell lysates, of which six (M-CSF, TREM-1, SICAM-1, JE, IL1ra, and MIP2) had significant differences between genotypes. Paraquat exposure to macrophages did not induce any significant genotype changes. M-CSF was the only pro-inflammatory cytokine that showed a significant difference between macrophages incubated with control media (Fig. 3-7). Wildtype macrophages had a greater presence of M-CSF compared to the H67D macrophages. Five different pro-inflammatory cytokines had significant differences in macrophages exposed with
LPS. Pro-inflammatory cytokines, TREM-1 and SICAM-1, were both significantly increased within wildtype macrophages, whereas JE, IL-1ra, and MIP-2 were significantly increased in
H67D macrophages (Fig. 3-7).
69 Figure 3-7 M-CSF TREM-1 1.0 0.15 ** Control Control 0.8 LPS LPS Paraquat 0.10 * Paraquat 0.6
0.4 0.05 Fold Change Fold Change 0.2
0.0 0.00
Wt Wt H67D H67D
JE SICAM-1 0.4 1.5 Control Control **** LPS LPS 0.3 *** Paraquat 1.0 Paraquat
0.2
0.5 Fold Change 0.1 Fold Change
0.0 0.0
Wt Wt H67D H67D
Il-1ra MIP-2 1.5 0.4 * Control Control LPS LPS 0.3 **** 1.0 Paraquat Paraquat
0.2
0.5 Fold Change Fold Change 0.1
0.0 0.0
Wt Wt H67D H67D
Figure 3-7: Pro-inflammatory Cytokine Array Panel.
Cell lysates from macrophages conditioned with control media, 50 ng/ml LPS, or 100 uM paraquat were run on a nitrocellulose membrane containing 40 different pro-inflammatory cytokines. Of which only six had readings above baseline. The membrane was exposed using GE
Amersham Imager 600. Data represent the mean ± SEM from three independent experiments and are compared to control for statistical significance using two-way ANOVA. *p<0.05, **p<0.01,
***p<0.001. ****p<0.0001.
70 3.5 Discussion
The key question addressed in this study is whether macrophage genotype affects macrophage function. Macrophages play a critical role in iron homeostasis and immune response, through sequestering and recycling of iron, and contributing to host defense mechanisms in the body10-13.
The HFE genotype is of particular importance because it impacts iron homeostasis which is a major function of macrophages32. The results of this study demonstrate that mutations in the HFE genotype significantly impact a number of critical macrophage functions, but differences in iron handling are minimal in the “resting” macrophages. A number of genotype specific responses were identified in this study; including proliferation and L- ferritin expression in response to iron exposure and secretion of BMP6 and cytokines. In addition to the HFE mutant specific changes to cytokine secretion profiles, two significant components of the immune response, migration and phagocytosis, were affected by changes in genotype as well. A significant observation, although not genotype specific, was the demonstration that exposure to apo-Tf (iron poor transferrin) can increase the release of iron from macrophages.
We first determined whether our treatment conditions affected cellular proliferation and survival in a genotype specific manner. WT macrophages treated with iron demonstrated a significant increase in cellular proliferation suggesting differences in iron utilization between wildtype and
H67D macrophages. Key iron management proteins including L-ferritin, H-ferritin, transferrin receptor, and ferroportin were then analyzed to further elucidate the differences of iron regulation due to genotype. In addition, macrophages labeled with 59Fe were exposed to potential iron chelators to determine the role of HFE in cellular iron release. Both the wildtype and H67D macrophages were able to load iron and store it in the form of L-ferritin but the H67D BMMs expressed significantly more L-ferritin than the wildtype BMMs (Fig. 3-2A). These data in combination with the proliferation analysis support the concept of differences in iron handling
71 because iron exposure increased proliferation in the WT cells but was associated with more iron storage in L-ferritin in the mutant cells. Interestingly, there was no significant difference in H- ferritin between the genotypes, a protein shown to be released from macrophages in a previous study33 (Figure 3-2B). The iron export protein, ferroportin, was also found to be significantly increased in the macrophages loaded with iron (Fig. 3-2C), however, no significant difference between genotypes was seen. There were no differences between any genotype or treatment group in reference to the iron import protein, transferrin receptor (TfR) (Fig. 3-2D). The regulation of ferroportin, ferritin and transferrin receptor is through iron regulatory element/iron regulatory protein system (IRE/IRP)34,35. This post-transcriptional regulation normally results in complimentary expression of proteins in response to iron changes. Thus, the lack of a predictable response in IRE/IRP regulated protein expression in macrophages is an area for further investigation.
BMMs have the ability to uptake iron and release it via ferroportin36 and H-ferritin37, but regulation of release of iron is not clear. To address iron release, and possibly gain insights into the iron depletion in macrophages in hemochromatosis, we analyzed iron release in our model.
Macrophages were incubated with control media, apo-Tf, DFO, or in combination with hepcidin following an overnight 59Fe incubation. First, we demonstrated that iron release was increased in the presence of apo-Tf and DFO (Fig. 3-4). This novel finding demonstrates for the first time that iron release from macrophages is signaled by the presence of apo-Tf and DFO. The physiological significance of this finding is that circulating transferrin in the serum, which is 70% unsaturated38, can serve as a mechanism to induce release of iron from macrophages. Moreover, given the response by the macrophages to release iron when exposed to DFO, it can be further argued that treatment with iron chelators in clinical settings can also remove iron from macrophages. Thus, there is a new and unappreciated function of apo-Tf uncovered in these studies and its potential
72 role in recycling iron from macrophages. These data do not however, help to explain the observation of iron poor macrophages in hemochromatosis because in this condition, Tf is reportedly highly saturated39. The presence of hepcidin in the media did not limit the iron release by DFO or apo-Tf indicating there is another mechanism for iron release that is not directly responsive to hepcidin. We have previously reported that apo-Tf can induce iron release from endothelial cells of the blood-brain-barrier, so the function of apo-Tf to remove iron from cells may be a significant and underappreciated function of this protein in iron handling40.
BMP6 and ALK3 levels were measured in cell lysates and cell culture media because of their role in affecting the expression of hepcidin and subsequently iron status. There were no significant differences in the levels of BMP6 and ALK3 in cell lysates, however, there was significantly more BMP6 secreted into H67D cell culture media (Figs. 3-3 and 3-4). The similar cellular levels of BMP6 coupled with the increased secretion indicates H67D macrophages are producing more
BMP6 than wildtype (Fig. 3-3B). These results indicate that there should be an increase in hepcidin production by the mutant macrophages that would be expected to lead to increased cellular iron levels in these cells but this was not observed in this study. Moreover, it is reported that hemochromatosis patients have less in their macrophages. Both of these results suggest a decrease in circulating hepcidin in association with a HFE mutation.
A cytokine that was increased in this study with HFE genotype that may influence hepcidin directly is Il-1 receptor antagonist (Il-1ra). This cytokine can reduce the production of hepcidin through blocking the Il-1/Il-6 induced transcription of hepcidin 41. Il-1 is the most significant cytokine to upregulate hepcidin production42. Therefore, increased levels of Il-1ra can decrease hepcidin production through inhibition of Il-1.
73 In addition to the contributions, to iron homeostasis, macrophages play a key role as inflammatory cells, and are critical in the innate immune response. They possess the ability to migrate to different infection sites and engulf foreign particles. The H67D macrophages have increased phagocytic ability compared to wildtype (Fig. 3-6). However, there is an established literature that suggests the HFE genotype can affect the immune response as HFE knock out mice have an attenuated immune response43. There is an increased susceptibility to bacterial infections in hemochromatosis patients that can be attributed to greater iron availability associated with the decreased sequestration of iron in macrophages 44,45. The results of a chemotaxis migration assay showed H67D macrophages had slower migration rate than wildtype macrophages (Fig. 3-5).
This data suggests that the attenuated immune response in HFE mutants may be due in part to decreased macrophage migration. Moreover, the decrease in cytokines such as TREM-1 and sICAM-1 would be consistent with a less potent inflammatory response in the presence of the mutant HFE. Lastly, two cytokines involved in chemotaxis, macrophage inflammatory protein 2
(MIP-2) and JE were elevated in the HFE mutant macrophages. The increases in JE and MIP-2, are consistent with our previous report that monocyte chemoattractant protein-1 (MCP-1) is elevated in H63D neuroblastoma cells, and hemochromatosis patients also reported have an increase in MCP-146,47. These findings indicate that despite their limited capacity for migration, the HFE mutation may promote increased migration of lymphocytes. All other cytokines analyzed were decreased in comparison to WT, supporting the decreased immune response seen in in HFE mutant population 47.
Together, these results suggest that changes in HFE genotype have major implications on macrophage phenotype and cause a significant alteration in macrophage functionality. These studies identify areas for further studies into HFE impact via macrophage dysfunction in the
74 context of neurodegenerative disorders such as Alzheimer's, ALS, and Parkinson’s, in addition to all types of cancer.
75 3.6 References
1. Wang J, Pantopoulos K. Regulation of cellular iron metabolism. The Biochemical
journal. 2011;434(3):365-381.
2. Feder JN, Gnirke A, Thomas W, et al. A novel MHC class I-like gene is mutated in
patients with hereditary haemochromatosis. Nat Genet. 1996;13(4):399-408.
3. Lebrón JA, Bennett MJ, Vaughn DE, et al. Crystal Structure of the Hemochromatosis
Protein HFE and Characterization of Its Interaction with Transferrin Receptor. Cell.
1998;93(1):111-123.
4. Bennett MJ, Lebron JA, Bjorkman PJ. Crystal structure of the hereditary
haemochromatosis protein HFE complexed with transferrin receptor. Nature.
2000;403(6765):46-53.
5. Steinberg KK, Cogswell ME, Chang JC, et al. Prevalence of C282Y and H63D Mutations
in the Hemochromatosis (HFE) Gene in the United States. JAMA. 2001;285(17):2216-
2222.
6. Burt MJ, George PM, Upton JD, et al. The significance of haemochromatosis gene
mutations in the general population: implications for screening. Gut. 1998;43(6):830-836.
7. Marshall DS, Linfert DR, Tsongalis GJ. Prevalence of the C282Y and H63D
polymorphisms in a multi-ethnic control population. International journal of molecular
medicine. 1999;4(4):389-393.
8. Valberg LS, Simon JB, Manley PN, Corbett WE, Ludwig J. Distribution of storage iron
as body iron stores expand in patients with hemochromatosis. Translational
Research.86(3):479-489.
9. Brink B, Disler P, Lynch S, Jacobs P, Charlton R, Bothwell T. Patterns of iron storage in
dietary iron overload and idiopathic hemochromatosis. The Journal of laboratory and
clinical medicine. 1976;88(5):725-731.
76 10. Zhou XY, Tomatsu S, Fleming RE, et al. HFE gene knockout produces mouse model of
hereditary hemochromatosis. Proceedings of the National Academy of Sciences.
1998;95(5):2492-2497.
11. Goodman GT, Koprowski H. Macrophages as a cellular expression of inherited natural
resistance. Proceedings of the National Academy of Sciences of the United States of
America. 1962;48(2):160-165.
12. Mackaness GB. Cellular immunity and the parasite. Advances in experimental medicine
and biology. 1977;93:65-73.
13. Gratchev A, Guillot P, Hakiy N, et al. Alternatively activated macrophages differentially
express fibronectin and its splice variants and the extracellular matrix protein betaIG-H3.
Scandinavian journal of immunology. 2001;53(4):386-392.
14. Montosi G, Paglia P, Garuti C, et al. Wild-type HFE protein normalizes transferrin iron
accumulation in macrophages from subjects with hereditary hemochromatosis. Blood.
2000;96(3):1125-1129.
15. Drakesmith H, Sweetland E, Schimanski L, et al. The hemochromatosis protein HFE
inhibits iron export from macrophages. Proceedings of the National Academy of
Sciences. 2002;99(24):15602-15607.
16. Wu X-g, Wang Y, Wu Q, et al. HFE interacts with the BMP type I receptor ALK3 to
regulate hepcidin expression. Blood. 2014;124(8):1335-1343.
17. Nicolas G, Bennoun M, Devaux I, et al. Lack of hepcidin gene expression and severe
tissue iron overload in upstream stimulatory factor 2 (USF2) knockout mice. Proceedings
of the National Academy of Sciences of the United States of America. 2001;98(15):8780-
8785.
77 18. Wang X-S, Lee S, Simmons Z, et al. Increased incidence of the Hfe mutation in
amyotrophic lateral sclerosis and related cellular consequences. Journal of the
Neurological Sciences. 2004;227(1):27-33.
19. Goodall EF, Greenway MJ, van Marion I, Carroll CB, Hardiman O, Morrison KE.
Association of the H63D polymorphism in the hemochromatosis gene with sporadic
ALS. Neurology. 2005;65(6):934-937.
20. Moalem S, Percy ME, Andrews DF, et al. Are hereditary hemochromatosis mutations
involved in Alzheimer disease? American journal of medical genetics. 2000;93(1):58-66.
21. Blázquez L, De Juan D, Ruiz-Martínez J, et al. Genes related to iron metabolism and
susceptibility to Alzheimer's disease in Basque population. Neurobiology of Aging.
2007;28(12):1941-1943.
22. Dekker MC, Giesbergen PC, Njajou OT, et al. Mutations in the hemochromatosis gene
(HFE), Parkinson's disease and parkinsonism. Neurosci Lett. 2003;348(2):117-119.
23. Guerreiro R, Bras J, Santana I, et al. Association of HFE common mutations with
Parkinson's disease, Alzheimer's disease and mild cognitive impairment in a Portuguese
cohort. BMC Neurology. 2006;6(1):24.
24. Lull ME, Block ML. Microglial Activation & Chronic Neurodegeneration.
Neurotherapeutics : the journal of the American Society for Experimental
NeuroTherapeutics. 2010;7(4):354-365.
25. Gunel-Ozcan A, Alyılmaz-Bekmez S, Guler EN, Guc D. HFE H63D mutation frequency
shows an increase in Turkish women with breast cancer. BMC Cancer. 2006;6:37-37.
26. Gharib AF, Karam RA, Pasha HF, Radwan MI, Elsawy WH. Polymorphisms of
hemochromatosis, and alpha-1 antitrypsin genes in Egyptian HCV patients with and
without hepatocellular carcinoma. Gene. 2011;489(2):98-102.
78 27. Guruvayoorappan C. Tumor Versus Tumor-Associated Macrophages: How Hot is the
Link? Integrative Cancer Therapies. 2008;7(2):90-95.
28. Kzhyshkowska J, Riabov V, Gudima A, Wang N, Orekhov A, Mickley A. Role of tumor
associated macrophages in tumor angiogenesis and lymphangiogenesis. Frontiers in
Physiology. 2014;5(75).
29. Tomatsu S, Orii KO, Fleming RE, et al. Contribution of the H63D mutation in HFE to
murine hereditary hemochromatosis. Proceedings of the National Academy of Sciences.
2003;100(26):15788-15793.
30. Zhang X, Goncalves R, Mosser DM. The Isolation and Characterization of Murine
Macrophages. Current protocols in immunology / edited by John E. Coligan ... [et al.].
2008;CHAPTER:Unit-14.11.
31. Bali PK, Zak O, Aisen P. A new role for the transferrin receptor in the release of iron
from transferrin. Biochemistry. 1991;30(2):324-328.
32. Makui H, Soares RJ, Jiang W, Constante M, Santos MM. Contribution of Hfe expression
in macrophages to the regulation of hepatic hepcidin levels and iron loading. Blood.
2005;106(6):2189-2195.
33. Connor JR, Zhang X, Nixon AM, Webb B, Perno JR. Comparative Evaluation of
Nephrotoxicity and Management by Macrophages of Intravenous Pharmaceutical Iron
Formulations. PLOS ONE. 2015;10(5):e0125272.
34. Hentze MW, Caughman SW, Rouault TA, et al. Identification of the Iron-Responsive
Element for the Translational Regulation of Human Ferritin mRNA. Science.
1987;238(4833):1570-1573.
35. Hentze MW, Caughman SW, Casey JL, et al. A model for the structure and functions of
iron-responsive elements. Gene. 1988;72(1–2):201-208.
79 36. Delaby C, Pilard N, Puy H, Canonne-Hergaux F. Sequential regulation of ferroportin
expression after erythrophagocytosis in murine macrophages: early mRNA induction by
haem, followed by iron-dependent protein expression. Biochemical Journal.
2008;411(1):123.
37. Cohen LA, Gutierrez L, Weiss A, et al. Serum ferritin is derived primarily from
macrophages through a nonclassical secretory pathway. Blood. 2010;116(9):1574-1584.
38. Luck AN, Mason AB. Transferrin-mediated cellular iron delivery. Current topics in
membranes. 2012;69:3-35.
39. Lee G, J F, Lukens J, et al. Hemochromatosis. In: Lee G, J F, J L, F P, JP G, GM R, eds.
Wintrobe’s Clinical Hematology. Vol 1. 10 ed. Baltimore, MD: Williams and Wilkens;
1999:1056-1070.
40. Simpson IA, Ponnuru P, Klinger ME, et al. A novel model for brain iron uptake:
introducing the concept of regulation. Journal of cerebral blood flow and metabolism :
official journal of the International Society of Cerebral Blood Flow and Metabolism.
2015;35(1):48-57.
41. Wrighting DM, Andrews NC. Interleukin-6 induces hepcidin expression through STAT3.
Blood. 2006;108(9):3204-3209.
42. Lee P, Peng H, Gelbart T, Wang L, Beutler E. Regulation of hepcidin transcription by
interleukin-1 and interleukin-6. Proceedings of the National Academy of Sciences of the
United States of America. 2005;102(6):1906-1910.
43. Wang L, Johnson EE, Shi HN, Allan Walker W, Wessling-Resnick M, Cherayil BJ.
Attenuated inflammatory responses in hemochromatosis reveal a role for iron in the
regulation of macrophage cytokine translation. Journal of immunology (Baltimore, Md. :
1950). 2008;181(4):2723-2731.
80 44. Gerhard GS, Levin KA, Price Goldstein J, Wojnar MM, Chorney MJ, Belchis DA. Vibrio
vulnificus septicemia in a patient with the hemochromatosis HFE C282Y mutation.
Archives of pathology & laboratory medicine. 2001;125(8):1107-1109.
45. Bullen JJ, Spalding PB, Ward CG, Gutteridge JM. Hemochromatosis, iron and septicemia
caused by Vibrio vulnificus. Archives of internal medicine. 1991;151(8):1606-1609.
46. Mitchell RM, Lee SY, Randazzo WT, Simmons Z, Connor JR. Influence of HFE variants
and cellular iron on monocyte chemoattractant protein-1. Journal of Neuroinflammation.
2009;6:6-6.
47. Lawless MW, White M, Mankan AK, O'Dwyer MJ, Norris S. Elevated MCP-1 serum
levels are associated with the H63D mutation and not the C282Y mutation in hereditary
hemochromatosis. Tissue antigens. 2007;70(4):294-300.
Chapter 4
The Neuroprotective Role of the HFE Gene Variant H67D in a Paraquat Mouse Model
4.1 Abstract
Parkinson’s disease (PD) is marked clinically by motor dysfunction, pathologically by dopaminergic cell loss at substantia nigra. Growing literature demonstrates iron accumulation in the substantia nigra (SN) of PD patients. The exact role of iron in PD etiology, is, however, unknown. The HFE protein is critical in the regulation of cellular iron uptake. Mutations within this protein are associated with increased iron accumulation within parenchymal cells. To investigate the role of H63D (a common HFE mutation), in paraquat-medicate neurotoxicity, we generated a mouse model in which the wild-type (WT) HFE gene is replaced by the H67D gene variant (mouse homologue of the human H63D gene variant), and injected with paraquat. The
WT paraquat-treated mice had significantly more falls compared to the WT saline-treated control group. In contrast, there were no behavioral differences between the H67D paraquat- and saline- treated mice. In addition, MRI analysis demonstrated an increase in R2 relaxation rate within the substantia nigra of the WT paraquat-treated mice compared to their saline-treated counterpart.
The H67D mice exhibited no significant R2 rate changes between treatment groups. To determine the cause of the behavioral and imaging changes, immunohistochemistry was used to evaluate tyrosine hydroxylase levels, microglia, and astrocytes. We observed a significant decrease in tyrosine hydroxylase and increase in microglia within the substantia nigra of WT paraquat-treated mice compared to the control consistent with the decrease in motor function. In contrast, H67D mice showed no differences in tyrosine hydroxylase and glia between the treatment groups.
However, these mice had significantly fewer TH neurons than WT. There was an increase in L-
82 ferritin with the substantia nigra of H67D saline treated mice which could have contributed to the neuroprotection to the paraquat injections. Our results clearly demonstrate that the HFE genotype impacts the level of TH neurons in the substantia nigra and the response to paraquat suggesting the genotype could be disease modifier for PD.
4.2 Introduction
Iron plays a critical role in biological and physiological processes, including within the nervous system; in processes such as neurotransmission, neuronal metabolism, and myelination1,2.
Consequently, mismanagement of iron can result in significant effects on brain related diseases.
Accumulation of iron is reported in the aging brain, specifically within regions of the substantia nigra, globus pallidus, and red nucleus3. These elevated iron levels have been associated with
Parkinson’s Disease4. Parkinson’s Disease is a common neurodegenerative disease characterized by the loss of dopamine neurons located in the substantia nigra (SN)5. Specifically, higher iron content in Parkinson’s Diseases patients, have been demonstrated by both postmortem and MRI studies, is greatly intensified6-10. The relationship between the pathogenic process of the disease and iron is not well understood. The over-riding question is if the increase in iron is a by-stander effect or is driving PD. The study used the paraquat toxicity model for neurodegeneration of tyrosine hydroxylase neurons to monitor the relationship between motor effects following paraquat and changes in MRI. The choice of paraquat is also because the pesticide have been linked to sporadic PD in large scale human epidemiology study11,12.
In PD, increased levels of brain iron can lead to an elevated α-synuclein levels, which promote the formation of Lewy bodies13. Generation of Lewy bodies results in protein aggregates, a principal feature of Parkinson’s Disease13,14. Free iron within the brain can also lead to reactive oxidative species (ROS) and neuronal death, which can contribute to the pathogenesis of
83 Parkinson’s Disease15. Similarly, as in other neurodegenerative diseases, PD patients have an increase in microglia, which can be contributed to the increase in α-synuclein16. The degeneration of dopaminergic neurons may be mediated through the increase in iron from the production of oxidative stress and activated microglia17.
In addition, given the potential impact of iron on neurodegenerative diseases, there has been increasing research on the iron management protein HFE which is thought to regulate cellular iron uptake by binding to the transferrin receptor-118,19. Mutations within this protein results in hemochromatosis, systemic iron overload, including increased levels of iron within the brain20,21.
The two primary mutations that occur are C282Y and H63D22. The C282Y mutation is commonly found within hemochromatosis patients, whereas the H63D mutation is found less frequently in hemochromatosis but it is one of the most common mutations in the Caucasian population22-25.
HFE mutations lead to aberrant iron storage, have been reported as a disease modifier for neurodegenerative diseases such as Alzheimer’s Disease (AD), Amyotrophic Lateral Sclerosis
(ALS), and Parkinson’s Disease (PD) where iron accumulation has been reported as part of the pathological profile26.
The clinical association between HFE and PD is unclear as there are conflicting results. Studies have shown that the HFE mutation, C282Y has an increased frequency in PD, however, there are other studies that do not find a connection between the C282Y and H63D HFE mutations and
PD27-32. The lack of consistency in a relationship between HFE genotype and PD could be an example of gene-environment interaction. Iron must be obtained from the diet and the presence of HFE mutant proteins will increase the absorption relative to wildtype HFE33. HFE genotype also alters the inflammatory response, possibly through impacting iron status, which is considered part of the pathology in PD34. In this study, we developed a Parkinsonism model with the
84 herbicide paraquat to evaluate the neurological consequences of the HFE gene variant, H67D
(mouse homolog of the H63D mutation) in a controlled environmental setting.
4.3 Materials and Methods
4.3.1 Animals
C57BL/6J x 129 mice (3-month old males) that contained either wildtype (WT) or H67D alleles, a homolog for the human H63D HFE mutation, were used in this study. The generation of the
H67D mice has been previously described35. Mice were individually housed and given ad libitum access to rodent chow pellets and water. The mice were maintained in an in-house animal facility at The Pennsylvania State University College of Medicine, and all procedures were approved by the Pennsylvania State University College of Medicine Institutional Animal Care and Use Committee, protocol 46457. The study consisted of 24 animals, of which a subset of 20 animals were used for MRI and immunohistochemistry.
4.3.2 Genotyping
To confirm the genotype of the mice, a portion of the tail was clipped and processed using
DNeasy blood and tissue kit (QIAGEN, Valencia, CA). The extraction was amplified through
PCR using following forward and reverse primers:
(5′AGGACTCACTCTCTGGCAGCAGGAGGTAACCA3′) and
(5′TTTCTTTTACAAAGCTATATCCCCAGGGT3′), with the following conditions; 4 °C for
15 min, 94 °C for 45 s, 58 °C for 45 s, 39 cycles of 72 °C for 90 s and 72 °C for 10 min.
Following PCR, the amplified DNA was digested for 2 hours at 37 °C using the restriction enzyme BspHI. To confirm the presence, or lack thereof, of the H67D mutation DNA was run on a 1.5% TAE gel and imaged with GE Amersham Imager 600 (GE; Buckinghamshire, United
Kingdom).
85 4.3.3 Paraquat and Saline Injections
The mice were divided into two groups which received either intraperitoneal saline (control) or
10 mg/kg paraquat dichloride hydrate (Sigma Aldrich; St. Louis, MO) injections. Injections were given once a week for three weeks (Figure 4-1). Paraquat was chosen to mimic the dopaminergic cell loss in Parkinson’s Disease 36.
86 Figure 4-1 Behavior Behavior Test 1 Tr a ining (Day Before Injection) Behavior Test 2
Food & Body Food & Body Food & Body Food & Body Weight Weight Weight Weight Paraquat or Saline Injection Harvest tissue
MRI MRI
1 2 3 4 5 6 7
Time (weeks)
Figure 4-1: Schematic Timeline of the Study Design.The food and body weight of the mice were taken each week throughout the seven week study. Behavior training began at week two, in which behavioral tests were recorded on weeks three and six. Following the first behavior test, the mice were divided into groups and given paraquat or saline injections once a week for three weeks. Following the last behavior test, the animals were sacrificed and tissue was harvested.
87 4.3.4 Behavior
Motor performance was tested using a rotarod apparatus (Columbus Instruments; Columbus,
OH). The mice were initially trained on the rotarod for 180 seconds at 15 rpm. Motor performance was tested and recorded twice, prior to paraquat or saline injections and prior to the endpoint (Figure 4-1). All tests lasted for 180 seconds at 15 rpm The number of falls were recorded for each behavior test, in which the falls from the second test were subtracted from the first.
4.3.5 Histology
At week seven, mice (N=20 animals (5/g) from the MRI group) were transcardially perfused with
Ringer’s solution. Following, the brains were removed, hemisected, and preserved with 4% paraformaldehyde in 0.1 M phosphate buffer. Following 24-hours the paraformaldehyde was replaced with fresh paraformaldehyde for an additional 24 hours. Lastly, paraformaldehyde was replaced with 70% ethanol until paraffin embedded.
4.3.5.1 Immunohistochemistry
Paraffin embedded tissue were coronally sectioned at 5 um thick. The sections were then deparaffinized and rehydrated through a series of xylene, ethanol, and finally deionized water.
Rehydration was followed by antigen retrieval using sodium citrate (pH 6.0) and then hydrogen peroxide to block endogenous peroxidases (20 minutes at room temperature). The sections were then blocked in 2% milk in 1X PBS for an hour at room temperature, followed by overnight incubation with 1:1000 tyrosine hydroxylase (Pel-Freez Biologicals; Rogers, AK), 1:1000 L- ferritin (Abcam; Cambridge, MA), 1:1000 IBA-1 (Wako; Richmond, VA), or 1:1000 GFAP
(Dako; Carpinteria, CA). The sections were washed and then incubated with 1:200 biotinylated anti-host secondary antibody (Vector Laboratories; Burlingame, CA) for one-hour at room
88 temperature. Immunoreactivity was detected using the avidin biotin complex (ABC) and 3,3′- diaminobenzidine (DAB) (Vector Laboratories; Burlingame, CA). The sections were imaged using the whole slide scanner, Aperio AT, and quantitatively analyzed using ImageJ software. images. DAB-induced staining intensity was quantified within a constant region of interest (ROI) surrounding the substantia nigra across all samples. Signal above a constant threshold was used to calculate the area positive for DAB within the ROI (i.e. area fraction).
4.3.5.2 Iron Staining
A modified Perl’s stain was used to assess iron within tissue sections37. Five µm paraffin embedded sections were deparaffinized and rehydrated using SafeClear Xylene Substitute (2 x 3 min; Thermo Fisher Scientific, Waltham, MA). Slides were then air dried for approximately 30 minutes. To stain for iron, slides were incubated with 1% potassium ferrocyanide trihydrate, 5%
PVP, and 0.05N HCl solution for 60 minutes at room temperature. Slides were quickly rinsed twice in water, and then blocked using methanol containing 0.3% hydrogen peroxide for 75 minutes. Next, the slides were washed twice in 1X PBS for 5 minutes. The slides were then placed into SafeClear Xylene Substitute prior to coverslipping with Cytoseal-60 mounting media
(Thermo Fisher Scientific, Waltham, MA). The sections were imaged using the whole slide scanner, Aperio AT, and quantitatively analyzed using ImageJ software as stated in the immunohistochemistry subsection.
4.3.6 Magnetic Resonance Imaging (MRI)
4.3.6.1 Image Acquisition
A subset of 20 animals (N=5/group, H67D, WT, paraquat, and saline) were anesthetized with
1.5% isoflurane, placed within a 35 mm birdcage volume coil using a standardized animal bed, and imaged with a 7.0 T Bruker BioSpec 70/20 MRI system (Bruker BioSpin, Ettlingen,
89 Germany). Animals were imaged with the same protocol at baseline prior to paraquat exposure and three-weeks later following paraquat treatment. A multi-echo three-dimensional RARE spin- echo T2-weighted protocol was utilized with the following parameters: averages = 2, relaxation time (TR) = 2000 ms, rare-factor = 8, four echoes with effective echo time (TE) = 30 – 120 ms with an echo spacing of 30 ms, field of view (FOV) = 25 x 25 x 15 mm, and acquisition matrix =
256 x 256 x 32, for a final voxel resolution of 97 x 97 x 468 µm.
4.3.6.2 MRI Parametric Analysis
Transverse relaxation rate (R2) maps were generated using a nonlinear least squares curve fitting model on a pixel-by-pixel basis 219,220 with qMRI software running in IDL 8.1. The first echo from each dataset was then skull-stripped with in-house software in SPM8 (Wellcome Trust
Centre for Neuroimaging, UK) and applied to all remaining echoes before spatial processing. The skull stripped first echo was then used as an anatomical image to spatially realign and co-register the parametric maps to a template mouse brain 221. Anatomical images and parameter maps were then normalized and resliced to the Magnetic Resonance Microimaging Neurological Atlas
(MRM NeAt ) template mouse brain 221,222 with a voxel size of 100 x 100 x 100 µm using SPM8 and the SPMmouse v1.1b toolkit 223, followed by smoothing with a 400 µm isotropic Gaussian smoothing kernel.
4.3.7 Statistical Analysis
Statistical analyses for behavior data and histologically sections were analyzed using the
Graphpad Prism (La Jolla, CA), in which the results were presented as mean ± SD. Statistical comparisons were completed by either a ungrouped t-test or a one-way ANOVA. A p-value of
≤0.05 was considered statistically significant.
90 MRI statistical analyses consisted of group based statistical parametric two-sample T-tests between the WT, H67D, paraquat, and saline treated groups at baseline and three-week time points using SPMmouse. Group based interaction effects between H67D and WT genetics over time were performed on the normalized R2 parametric maps using a modified SPM contrast matrix with an absolute threshold for voxel cluster ≥ 50 in size with p value ≤ 0.005. Outliers were excluded if they were statistically different from the other values.
4.4 Results
A schematic representing drug administration and behavior analysis is represented in Figure 4-1.
Briefly, over the course of seven weeks the amount of food consumed and body weight was measured every week and in which there was no significant differences between genotype and treatment groups (Figure 4-2A and 4-2B). Behavior and motor function was assessed prior and following paraquat and saline injections through rotarod apparatus and MRI analysis (Figures 4-3 and 4-4). Following the last rotarod test, the mice were sacrificed and brain tissue was harvested for immunohistochemistry (Figures 4-5 through 4-9).
91 Figure 4-2 45 4 A. B. 40 2 0 35 -2 30 -4
25 -6 Change in Body Weight (g) Mean Mean Food Intake (kcal/kg) 20 -8
WT Saline WT Saline H67D Saline WT Paraquat H67D Saline WT Paraquat H767D Paraquat H767D Paraquat
Figure 4-2: Mean Food Intake and Change in Body Weight. To ensure treatment or genotype did not impact food intake (A) or body weight (B), mean food intake and change in body weight was measured. Food was weighed out at the beginning and end of each week. To determine food intake, the weights were subtracted and divided by seven days to get the dailey food intake. At the same time points and in a similar manner, each mouse was weighed and change in body weight was assessed. There was no significant differnces of mean food intake or change in body weight across treatment/genotype groups. Data represent the mean ± SD and are compared to all treatment groups for statistical significance using 1-way ANOVA. (N=5 or 6/group)
92 4.4.1 Effect of paraquat on motor function
The loss of motor function, induced by paraquat, was analyzed through the use of a rotarod apparatus and MRI. Following three weeks of paraquat or saline injections, wildtype mice receiving paraquat injections had a greater number of falls (p<0.05) compared to the wildtype saline injected mice, indicating a greater motor loss (Figure 4-3). However, there were no significant differences in the number of falls between H67D mice receiving paraquat or saline injections (Figure 4-3). MRI analysis revealed a longitudinal increase in R2 relaxation rate for the paraquat injected wildtype mice compared to the wildtype saline injected mice, specifically within the regions of the substantia nigra, zona incerta, and ventral midbrain (Figure 4-4 top). In contrast, there were no significant changes in R2 relaxation rate between H67D paraquat and saline injected mice (Figure 4-4 bottom).
Immunohistochemistry (Figure 4-5 through 4-9) was performed on the animals at the end of the study to identify the histopathological changes that may underlie the behavioral motor loss and
MRI differences in paraquat injected wildtype mice, in addition to the lack of changes seen within the H67D mice.
93 Figure 4-3 Data 6 * 5
0
-5 Number of Falls
-10
Wt Saline H67D Saline Wt Paraquat H767D Paraquat
Figure 4-3: Behavior Analysis of Motor Function between Genotype and Treatment
Groups. Motor performance was assessed using a rotarod, in which the number of falls were from the two behavior tests were subtracted from each other. An increase in the number of falls indicates a decrease in motor performance and increase in disease progression. The wildtype paraquat injected mice had an increase in number of falls compared to their saline counterpart.
Whereas, there was no significant difference in the number of falls between the H67D treatment groups. Data represent the mean ± SD and are compared to all treatment groups for statistical significance using 1-way ANOVA. *p<0.05; N=5 or 6/group
94 Figure 4-4
Figure 4-4: MRI Analysis of Genotype and Treatment Groups. Following behavior tests, mice underwent MRI to determine any correlation between motor function and cellular changes represented by R2 relaxation rate. Comparisons between the genotypes and treatment groups showed a longitudinal increase in R2 rate in wildtype mice injected with paraquat within substantia nigra, zona incerta, and ventral midbrain compared to the wildtype saline mice (top).
However, there was no longitudinal R2 rate change observed between H67D paraquat and saline mice (bottom). Images are presented with a voxel threshold of p < 0.005 with a minimal cluster size of 50.
95 4.4.2 Tyrosine hydroxylase neuron loss
The effect of paraquat treatment on tyrosine hydroxylase (TH) neurons within the substantia nigra pars compacta revealed a 58% loss of TH neurons in the pars compacta of the wildtype injected paraquat mice compared to the wildtype saline mice (Figure 4-5B and 4-5D; p<0.05). The H67D mice have 71.5% less TH neurons than wildtype mice (Figure 4-5A and 4-5C; p<0.01 and p<0.05). There is no significant difference in TH neuronal loss between H67D paraquat and saline injected groups.
96 Figure 4-5 H67D Wildtype A B Saline
C D Paraquat
97
TH Neurons 15 * * H67D Saline ** H67D Paraquat 10 WT Saline WT Paraquat
5 % Area Fraction
0
WT Saline H67D Saline WT Paraquat H67D Paraquat
Figure 4-5: Tyrosine Hydroxylase (TH) Expression in the Substantia Nigra. Coronal sections were incubated with 1:1000 anti-TH antibody. Results show wildtype paraquat injected mice (D) have a significant decrease of tyrosine hydroxylase neurons within the substantia nigra compared to wildtype saline mice (B). Furthermore, there is a significant loss of TH neurons within the substantia nigra of the H67D saline mice (A) and H67D paraquat mice (C) compared to wildtype saline (B). Data represent the mean ± SD and are compared to all treatment groups for statistical significance using 1-way ANOVA. *p<0.05, **p<0.01; N=5/group
98 4.4.3 L-ferritin levels within the substantia nigra
Next, we evaluated differences in the long term iron storage protein, L-ferritin. The cells that stained positive for L-ferritin resembled microglia and oligodendrocytes, regardless of genotype or treatment (Figure 4-6A). In the pars compacta, quantitative analysis revealed that H67D saline injected mice have approximately twice the amount of L-ferritin compared to H67D paraquat injected mice (Figure 4-6B; p<0.05). In addition, the amount of L-ferritin in the H67D saline injected mice was significantly greater than wildtype injected saline and paraquat mice by approximately five times the amount (Figure 4-6B; p<0.01). Paraquat injections had no effect on
L-ferritin in the wildtype mice. There were no significant differences found across genotype and treatment groups within the pars reticulata (Figure 4-6C).
99 Figure 4-6
A
B L-ferritin (Pars Compacta) C L-ferritin (Pars Reticulata) 100 ** 80 ** H67D Saline H67D Saline
80 * H67D Paraquat H67D Paraquat 60 Wildtype Saline Wildtype Saline 60 Wildtype Paraquat Wildtype Paraquat 40 40 20 % Area Fraction % Area Fraction 20
0 0
H67D Saline H67D Saline H67D ParaquatWildtype Saline H67D ParaquatWildtype Saline Wildtype Paraquat Wildtype Paraquat
100 Figure 4-6: L-ferritin Expression in the Substantia Nigra. Coronal sections were incubated with 1:1000 anti-L-ferritin antibody. A representative slide shows the accumulation of L-ferritin within microglia and oligodendrocytes (A). Results show there is significantly more L-ferritin present within the substantia nigra pars compacta of H67D saline mice compared to H67D paraquat mice and wildtype saline mice (B). There were no differences in L-ferritin expression within the substantia nigra pars reticulata (C). Data represent the mean ± SD and are compared to all treatment groups for statistical significance using 1-way ANOVA. *p<0.05, **p<0.01;
N=5/group
101 4.4.4 Microglia and Astrocyte Changes within Substantia Nigra
To interrogate the role of gliosis in these models, we examined markers of astrocytes and microglia. In wildtype mice with paraquat injections there was a 3-fold increase in microgliosis within the pars compacta of the substantia nigra compared to the saline injected wildtype mice, however the difference was not statistically significant (Figure 4-7A). In contrast, there was statistically significant microgliosis within the pars reticulata of the wildtype paraquat injected mice, in which there was four times the amount of microglia present, compared to the wildtype saline mice. The wildtype paraquat injected mice had about four times the amount of microglia compared to the H67D paraquat injected mice also within the pars reticulata (Figure 4-7B; p<0.01). Despite the increase in microgliosis, there was no astrogliosis observed with in the wildtype paraquat injected mice in either the pars compacta or pars reticulata (Figure 4-8A and 4-
8B). Furthermore, there was no significant astrogliosis present in the H67D mice within both regions of the substantia nigra (Figure 4-8A and 4-8B).
102 Figure 4-7
IBA1 (Pars Reticulata) A IBA1 (Pars Compacta) B IBA1 (Pars Compacta)Reticulata) 2010 1020 H67D Saline H67D Saline 8 H67D8 Paraquat H67D Paraquat 15 15 Wildtype Saline Wildtype Saline 6 Wildtype6 Paraquat ** Wildtype Paraquat 10 10 ** 4 4
5 5 % Area Fraction % Area Fraction % Area Fraction 2 % Area Fraction 2
0 0 0
H67DH67D Saline Saline H67D Saline H67D Paraquat H67D Paraquat H67D ParaquatWildtypeWildtype Saline Saline H67D ParaquatWildtype Saline WildtypeWildtype Paraquat Paraquat Wildtype Paraquat
Figure 4-7: Microglia Expression in the Substantia Nigra. Coronal sections were incubated with 1:1000 anti-IBA1 antibody. Results show there is significantly more microglia present within the substantia nigra pars reticulata of wildtype paraquat injected mice compared to wildtype saline mice and H67D paraquat mice (B). There were no differences in microglial expression within the substantia nigra pars compacta (A). Data represent the mean ± SD and are compared to all treatment groups for statistical significance using 1-way ANOVA. ** p<0.0;
N=5/group
103 Figure 4-8
A GFAP (Pars Compacta) B GFAP (Pars Reticulata) 15 20 H67D Saline H67D Saline H67D Paraquat H67D Paraquat 15 10 Wildtype Saline Wildtype Saline Wildtype Paraquat Wildtype Paraquat 10
5 5 % Area Fraction % Area Fraction
0 0
H67D Saline H67D Saline H67D ParaquatWildtype Saline H67D ParaquatWildtype Saline Wildtype Paraquat Wildtype Paraquat Figure 4-8: Astrocyte Expression in the Substantia Nigra. Coronal sections were incubated with 1:1000 anti-GFAP antibody. There were no differences in astroglial expression within the substantia nigra pars compacta (A) and pars reticulata (B), between any treatment or genotype groups. Data represent the mean ± SD and are compared to all treatment groups for statistical significance using 1-way ANOVA. (N=5/group)
104 4.4.5 Iron accumulation within the substantia nigra
Substantia nigra sections were stained with a modified Perl stain. At the cellular level, analysis showed that there were no significant differences in iron accumulation within the pars reticulata of the substantia nigra (Figure 4-9B). In addition, there were no significant differences between genotype or treatment groups (Figure 4-9B). However, the accumulation of iron resided within oligodendrocytes (Figure 4-9A).
105 Figure 4-9
Perl A B20 H67D Saline H67D Paraquat 15 Wildtype Saline Wildtype Paraquat 10
5 % Area Fraction
0
H67D Saline H67D ParaquatWildtype Saline Wildtype Paraquat
Figure 4-9: Iron (Perl) Expression in the Substantia Nigra. Coronal sections were incubated with potassium ferrocyanide trihydrate for iron staining. A representative slide shows the accumulation of iron oligodendrocytes (A). There were no significant differences in iron expression within the substantia nigra pars compacta and pars reticulate, between any treatment or genotype groups. Data represent the mean ± SD and are compared to all treatment groups for statistical significance using 1-way ANOVA. (n=5/group)
106 4.5 Discussion
In this study, we demonstrate that wildtype mice injected with paraquat have motor impairment, changes in MRI that reflect alterations in regional cellularity, loss of TH positive neurons and increased microgliosis. However, paraquat did not induce neurological changes within the H67D model, indicating the H67D HFE gene variant is neuroprotective against paraquat.
The HFE mutant, H67D, did not show any significant differences in motor function or changes in
MRI following paraquat injection. These data suggest that the HFE mutation could be neuroprotective. Consistent with this interpretation is the lack of microgliosis in the HFE paraquat injected mice. The only histological difference within the mutants was an increase in L- ferritin found within the pars compacta of the saline group, which could be part of the protective milieu in the SN of these mice to paraquat by sequestering iron and preventing cellular damage.
As an example that ferritin is protective in a neurotoxicant model, Kaur et al. found transgenic overexpression of H-ferritin in TH neurons limited loss of these cells 43. The elevated levels of ferritin are consistent with an MRI study which demonstrated an increase in relaxation rate within the male H63D carrier brain44.
One of the pathological hallmarks of Parkinson’s Disease is the progressive degeneration of dopaminergic neurons within the pars compacta region of the substantia nigra, resulting in the loss of motor function and the progression of the disease5. Tyrosine hydroxylase is an enzyme that contributes to the formation of dopamine within the neurons in the substantia nigra and is used as a surrogate marker for the dopaminergic neurons45. The loss of tyrosine hydroxylase neurons within the wildtype mice injected with paraquat is presumed to have lost their motor function as the loss of motor control in PD patients is attributed to the loss of TH positive neurons46. Our study showed an intraperitoneal injection of paraquat leads to a decrease in
107 tyrosine hydroxylase within the substantia nigra of wildtype mice. A paraquat-mice model was used to mimic Parkinson’s Disease as it selectively targets and damages dopaminergic neurons specific to the substantia nigra, and link to PD epidemiologically47,48. The etiology of neurotoxicity induced by paraquat is not well understood but thought to be the result of oxidative stress49. In this study associated with the neuronal cell loss was an increase in microglia. It is reported that microglia release pro-inflammatory cytokines, such as tumor necrosis factor alpha
(TNFa) and interleukin-1b into the pars compacta region of the substantia nigra, resulting in oxidative stress and cellular damage50 to neurons. In addition, a study conducted by Purisai et al. concluded activation of microglia through NADPH-oxidase induction, by paraquat, was the key event leading to paraquat induced dopaminergic neuron loss51. When Purisai et al. used minocycline, an anti-inflammatory drug, following paraquat injections, there was no oxidative stress or loss of TH+ neurons observed51. Therefore, the increased microglia activation seen in wildtype mice injected with paraquat in our study is likely induced by paraquat and responsible for the dopaminergic neurons cell loss. In contrast, there was not significant astrocyte difference seen within the wildtype paraquat and saline injected groups. Because the paraquat injection paradigm that we used is an acute model, the lack of significant increase in astrocytes may suggest these cells are more involved in a chronic disease process.
One of the goals of this study was to determine the temporal sequence of changes in MRI relative to behavioral and histological changes. MRI analysis was performed twice, once at baseline and
a second time following three-week paraquat exposure prior to sacrifice. The increase in MRI R2 rate in wildtype animals demonstrate a regional change in cellular content52-56, in which the wildtype mice injected with paraquat demonstrate an increase in cellular loss compared to the wildtype mice receiving saline injections. The focal R2 increase within wildtype animals over the three-weeks course is interpreted as a loss of dopaminergic neurons and/or an increase in
108 microglia proliferation within the outlined region. It is interesting to note that the main MRI differences located in the left side of mice brain. This finding is very intriguing because the onset of PD motor symptoms is typically asymmetric57 and more often present on the right side of the body58. Most recently, it has been demonstrated that the pesticide farmers also have left more than right side MRI changes in the basal ganglia59.
Whereas the wildtype mice injected with paraquat depicted histological and behavioral changes consistent with previous reports, there were significant differences between WT and H67D mice.
Moreover, the control (saline injected) wildtype and H67D mice were also significantly different on a number of measures. The H67D mice had significantly fewer TH neurons at the beginning of the study than the wildtype mice. There was no change of tyrosine hydroxylase neuronal density within H67D mutant mice receiving paraquat injections suggesting that there is a neuroprotective milieu in the H67D mice. We cannot at this time discern if the baseline decrease in H67D TH neurons was from previous damage during development because of the gene variant or flawed development that resulted in fewer TH neurons. Our data suggest that the initial decrease in
H67D TH neuron number could result from a flaw in development, because there were no differences found within the H67D longitudinal MRI data that are indicative of ongoing inflammation nor was there evidence of gliosis. The H67D mice injected with saline had on average a higher number of falls than the wildtype saline injected mice, however the difference was not significant. Nonetheless, the observation would suggest some degree of underlying motor impairment in the mutant mice associated with the significant loss of neurons. The data however do suggest, as also observed in PD, that significant neuronal loss must occur prior to clinical presentation of the disease60. Moreover, the limited motor impairment in the HFE mice further suggests that the lack of TH neurons in this model is a developmental variant to which the
109 animal has adapted. Further studies are warranted and may shed light on basal ganglia development and motor controls.
In summary, we examined the effect of the H67D gene variant on neurodegeneration in a paraquat mouse model. We conclude the H67D genotype has a neuroprotective role against neuronal cell loss induced by paraquat. This effect may be mediated by the increased sequestration of L-ferritin, preventing oxidative stress, and allowing them to handle the paraquat.
These results give insight for the human population with the equivalent H63D mutation to be resistant to the neurotoxicity of environmental toxins such as paraquat, and have a decreased risk for Parkinson’s Disease.
4.6 Acknowledgements
The work is supported in part by the Pennsylvania Department of Health Tobacco CURE Funds and the Penn State Neuroscience Institute.
Thank you to Beth Neely for helping with the behavior study, Dr. Mark Meadowcroft and Carson
Purnell for MRI analysis, and Jean Copper with assistance in IHC data acquisition.
110 4.7 References
1. Yu GS, Steinkirchner TM, Rao GA, Larkin EC. Effect of prenatal iron deficiency on myelination in rat pups. The American journal of pathology. 1986;125(3):620-624.
2. Wigglesworth J, Baum H. Iron dependent enzymes in the brain. Brain iron: neurochemical and behavioural aspects New York: Taylor and Francis. 1988:25-66.
3. Bilgic B, Pfefferbaum A, Rohlfing T, Sullivan EV, Adalsteinsson E. MRI estimates of brain iron concentration in normal aging using quantitative susceptibility mapping. NeuroImage.
2012;59(3):2625-2635.
4. Loeffler DA, Connor JR, Juneau PL, et al. Transferrin and iron in normal, Alzheimer's disease, and Parkinson's disease brain regions. J Neurochem. 1995;65(2):710-724.
5. Hirsch EC. Biochemistry of Parkinson's disease with special reference to the dopaminergic systems. Molecular neurobiology. 1994;9(1-3):135-142.
6. Wang JY, Zhuang QQ, Zhu LB, et al. Meta-analysis of brain iron levels of Parkinson's disease patients determined by postmortem and MRI measurements. Sci Rep. 2016;6:36669.
7. Gorell JM, Ordidge RJ, Brown GG, Deniau JC, Buderer NM, Helpern JA. Increased iron- related MRI contrast in the substantia nigra in Parkinson's disease. Neurology. 1995;45(6):1138-
1143.
8. Dexter DT, Wells FR, Agid F, et al. Increased nigral iron content in postmortem parkinsonian brain. The Lancet. 1987;330(8569):1219-1220.
9. Riederer P, Sofic E, Rausch W-D, et al. Transition Metals, Ferritin, Glutathione, and
Ascorbic Acid in Parkinsonian Brains. Journal of Neurochemistry. 1989;52(2):515-520.
10. Du G, Liu T, Lewis MM, et al. Quantitative susceptibility mapping of the midbrain in
Parkinson's disease. Movement disorders : official journal of the Movement Disorder Society.
2016;31(3):317-324.
111 11. McCormack AL, Thiruchelvam M, Manning-Bog AB, et al. Environmental risk factors and Parkinson's disease: selective degeneration of nigral dopaminergic neurons caused by the herbicide paraquat. Neurobiol Dis. 2002;10(2):119-127.
12. Tanner CM, Kamel F, Ross GW, et al. Rotenone, paraquat, and Parkinson's disease.
Environmental health perspectives. 2011;119(6):866-872.
13. Spillantini MG, Schmidt ML, Lee VMY, Trojanowski JQ, Jakes R, Goedert M. [alpha]-
Synuclein in Lewy bodies. Nature. 1997;388(6645):839-840.
14. Forno LS. Concentric hyalin intraneuronal inclusions of lewy type in the brains of elderly persons (50 incidental cases): relationship to parkinsonism. Journal of the American Geriatrics
Society. 1969;17(6):557-575.
15. Kruszewski M. Labile iron pool: the main determinant of cellular response to oxidative stress. Mutation research. 2003;531(1-2):81-92.
16. Roodveldt C, Labrador-Garrido A, Gonzalez-Rey E, et al. Preconditioning of Microglia by α-Synuclein Strongly Affects the Response Induced by Toll-like Receptor (TLR) Stimulation.
PLoS ONE. 2013;8(11):e79160.
17. Gao HM, Kotzbauer PT, Uryu K, Leight S, Trojanowski JQ, Lee VM.
Neuroinflammation and oxidation/nitration of alpha-synuclein linked to dopaminergic neurodegeneration. The Journal of neuroscience : the official journal of the Society for
Neuroscience. 2008;28(30):7687-7698.
18. Lebrón JA, Bennett MJ, Vaughn DE, et al. Crystal Structure of the Hemochromatosis
Protein HFE and Characterization of Its Interaction with Transferrin Receptor. Cell.
1998;93(1):111-123.
19. Bennett MJ, Lebron JA, Bjorkman PJ. Crystal structure of the hereditary haemochromatosis protein HFE complexed with transferrin receptor. Nature. 2000;403(6765):46-
53.
112 20. Nielsen JE, Jensen LN, Krabbe K. Hereditary haemochromatosis: a case of iron accumulation in the basal ganglia associated with a parkinsonian syndrome. Journal of neurology, neurosurgery, and psychiatry. 1995;59(3):318-321.
21. Berg D, Hoggenmuller U, Hofmann E, et al. The basal ganglia in haemochromatosis.
Neuroradiology. 2000;42(1):9-13.
22. Feder JN, Gnirke A, Thomas W, et al. A novel MHC class I-like gene is mutated in patients with hereditary haemochromatosis. Nat Genet. 1996;13(4):399-408.
23. Steinberg KK, Cogswell ME, Chang JC, et al. Prevalence of C282Y and H63D Mutations in the Hemochromatosis (HFE) Gene in the United States. JAMA. 2001;285(17):2216-2222.
24. Burt MJ, George PM, Upton JD, et al. The significance of haemochromatosis gene mutations in the general population: implications for screening. Gut. 1998;43(6):830-836.
25. Marshall DS, Linfert DR, Tsongalis GJ. Prevalence of the C282Y and H63D polymorphisms in a multi-ethnic control population. International journal of molecular medicine.
1999;4(4):389-393.
26. Nandar W, Connor JR. HFE Gene Variants Affect Iron in the Brain. The Journal of
Nutrition. 2011;141(4):729S-739S.
27. Dekker MC, Giesbergen PC, Njajou OT, et al. Mutations in the hemochromatosis gene
(HFE), Parkinson's disease and parkinsonism. Neurosci Lett. 2003;348(2):117-119.
28. Guerreiro R, Bras J, Santana I, et al. Association of HFE common mutations with
Parkinson's disease, Alzheimer's disease and mild cognitive impairment in a Portuguese cohort.
BMC Neurology. 2006;6(1):24.
29. Buchanan DD, Silburn PA, Chalk JB, Le Couteur DG, Mellick GD. The Cys282Tyr polymorphism in the HFE gene in Australian Parkinson's disease patients. Neuroscience Letters.
2002;327(2):91-94.
113 30. Akbas N, Hochstrasser H, Deplazes J, et al. Screening for mutations of the HFE gene in
Parkinson's disease patients with hyperechogenicity of the substantia nigra. Neuroscience Letters.
2006;407(1):16-19.
31. Aamodt AH, Stovner LJ, Thorstensen K, Lydersen S, White LR, Aasly JO. Prevalence of haemochromatosis gene mutations in Parkinson’s disease. Journal of Neurology, Neurosurgery &
Psychiatry. 2007;78(3):315-317.
32. Borie C, Gasparini F, Verpillat P, et al. Association study between iron-related genes polymorphisms and Parkinson's disease. Journal of Neurology. 2002;249(7):801-804.
33. Ajioka RS, Levy JE, Andrews NC, Kushner JP. Regulation of iron absorption in Hfe mutant mice. Blood. 2002;100(4):1465-1469.
34. Wang L, Johnson EE, Shi HN, Allan Walker W, Wessling-Resnick M, Cherayil BJ.
Attenuated inflammatory responses in hemochromatosis reveal a role for iron in the regulation of macrophage cytokine translation. Journal of immunology (Baltimore, Md : 1950).
2008;181(4):2723-2731.
35. Tomatsu S, Orii KO, Fleming RE, et al. Contribution of the H63D mutation in HFE to murine hereditary hemochromatosis. Proceedings of the National Academy of Sciences.
2003;100(26):15788-15793.
36. Brooks AI, Chadwick CA, Gelbard HA, Cory-Slechta DA, Federoff HJ. Paraquat elicited neurobehavioral syndrome caused by dopaminergic neuron loss. Brain Research. 1999;823(1–
2):1-10.
37. Sands SA, Leung-Toung R, Wang Y, Connelly J, LeVine SM. Enhanced Histochemical
Detection of Iron in Paraffin Sections of Mouse Central Nervous System Tissue: Application in the APP/PS1 Mouse Model of Alzheimer’s Disease. ASN NEURO.
2016;8(5):1759091416670978.
114 38. Dardzinski BJ, Mosher TJ, Li S, Van Slyke MA, Smith MB. Spatial variation of T2 in human articular cartilage. Radiology. 1997;205(2):546-550.
39. Koff MF, Amrami KK, Felmlee JP, Kaufman KR. Bias of cartilage T2 values related to method of calculation. Magn Reson Imaging. 2008;26(9):1236-1243.
40. Ma Y, Smith D, Hof PR, et al. In Vivo 3D Digital Atlas Database of the Adult C57BL/6J
Mouse Brain by Magnetic Resonance Microscopy. Front Neuroanat. 2008;2:1.
41. Ma Y, Hof PR, Grant SC, et al. A three-dimensional digital atlas database of the adult
C57BL/6J mouse brain by magnetic resonance microscopy. Neuroscience. 2005;135(4):1203-
1215.
42. Sawiak SJ, Wood NI, Williams GB, Morton AJ, Carpenter TA. Voxel-based morphometry in the R6/2 transgenic mouse reveals differences between genotypes not seen with manual 2D morphometry. Neurobiol Dis. 2009;33(1):20-27.
43. Kaur D, Yantiri F, Rajagopalan S, et al. Genetic or Pharmacological Iron Chelation
Prevents MPTP-Induced Neurotoxicity In Vivo: A Novel Therapy for Parkinson's Disease.
Neuron. 2003;37(6):899-909.
44. Bartzokis G, Lu PH, Tishler TA, et al. Prevalent Iron Metabolism Gene Variants
Associated with Increased Brain Ferritin Iron in Healthy Older Men. Journal of Alzheimer's disease : JAD. 2010;20(1):333-341.
45. Daubner SC, Le T, Wang S. Tyrosine Hydroxylase and Regulation of Dopamine
Synthesis. Archives of biochemistry and biophysics. 2011;508(1):1-12.
46. Gelb DJ, Oliver E, Gilman S. Diagnostic criteria for Parkinson disease. Arch Neurol.
1999;56(1):33-39.
47. Desplats P, Patel P, Kosberg K, et al. Combined exposure to Maneb and Paraquat alters transcriptional regulation of neurogenesis-related genes in mice models of Parkinson’s disease.
Molecular Neurodegeneration. 2012;7:49-49.
115 48. Uversky VN. Neurotoxicant-induced animal models of Parkinson's disease: understanding the role of rotenone, maneb and paraquat in neurodegeneration. Cell and tissue research. 2004;318(1):225-241.
49. McCarthy S, Somayajulu M, Sikorska M, Borowy-Borowski H, Pandey S. Paraquat induces oxidative stress and neuronal cell death; neuroprotection by water-soluble Coenzyme
Q10. Toxicology and applied pharmacology. 2004;201(1):21-31.
50. Lu X, Bing G, Hagg T. Naloxone prevents microglia-induced degeneration of dopaminergic substantia nigra neurons in adult rats. Neuroscience. 2000;97(2):285-291.
51. Purisai MG, McCormack AL, Cumine S, Li J, Isla MZ, Di Monte DA. Microglial activation as a priming event leading to paraquat-induced dopaminergic cell degeneration.
Neurobiology of disease. 2007;25(2):392-400.
52. Koenig SH, Brown RD, 3rd, Spiller M, Lundbom N. Relaxometry of brain: why white matter appears bright in MRI. Magnetic resonance in medicine. 1990;14(3):482-495.
53. Kucharczyk W, Macdonald PM, Stanisz GJ, Henkelman RM. Relaxivity and magnetization transfer of white matter lipids at MR imaging: importance of cerebrosides and pH.
Radiology. 1994;192(2):521-529.
54. Kamman RL, Go KG, Muskiet FA, Stomp GP, Van Dijk P, Berendsen HJ. Proton spin relaxation studies of fatty tissue and cerebral white matter. Magnetic resonance imaging.
1984;2(3):211-220.
55. Ranade SS. Proton relaxation in vivo, in vitro and cellularity. Acta radiologica
(Stockholm, Sweden : 1987). 1988;29(1):143.
56. Abdolmohammadi J, Shafiee M, Faeghi F, et al. Determination of intra-axial brain tumors cellularity through the analysis of T2 Relaxation time of brain tumors before surgery using MATLAB software. Electronic physician. 2016;8(8):2726-2732.
116 57. Riederer P, Sian-Hülsmann J. The significance of neuronal lateralisation in Parkinson’s disease. Journal of Neural Transmission. 2012;119(8):953-962.
58. Uitti RJ, Baba Y, Whaley NR, Wszolek ZK, Putzke JD. Parkinson disease: handedness predicts asymmetry. Neurology. 2005;64(11):1925-1930.
59. Lewis MM, Sterling NW, Du G, et al. Lateralized basal ganglia vulnerability to pesticide expsoure in asymptomatic agricultural workers. Toxicological Sciences. 2017.
60. Marsden CD. Parkinson's disease. The Lancet. 1990;335(8695):948-949.
117
Chapter 5
The Effect of Cholesterol and Statin Diet on Motor Function in H67D Paraquat Mouse Model
5.1 Abstract
The degeneration of dopaminergic neurons is the primary underlying mechanism of Parkinson’s
Disease, and has been the focus of research investigating the biology and therapeutic development of Parkinson’s. However, several other factors have been found to significantly affect the progression of Parkinson’s, such as cholesterol and iron dysregulation, and statin treatment. Through the use of the neurotoxin paraquat, and transgenic mice with the H67D HFE mutation, we were able to generate a model to study the effect of these factors on the neurodegeneration leading to impaired motor function. We found that mice with the H67D genotype had no motor impairment following injection with paraquat. In addition, cholesterol or statin diet did not affect behavior when injected with saline or paraquat in these mice. However, motor function of wildtype mice was impaired when injected with paraquat and a diet rich in cholesterol accelerates the effects of paraquat. There was no effect of statin exposure. These results indicate the HFE genotype, the most prevalent gene variant in Caucasians, affects the outcome in models of Parkinson’s disease.
5.2 Introduction
Parkinson’s Disease is the second most common neurodegenerative disease, pathologically represented by the loss of dopaminergic neurons located in the substantia nigra and the accumulation of the protein a-synuclein1,2. The reasons for the loss of neurons and accumulation
118 of proteins is not well known but thought to involve oxidative stress and inflammation resulting in neurotoxicity3. Potential factors that may contribute to neurotoxicity are cholesterol levels, statin treatment, and iron dysregulation.
Diets high in cholesterol have long been studied in regards to their potential link to cardiovascular disease. However, an increasing number of studies are now looking at cholesterol in the brain, as the brain is the most cholesterol rich organ4. Cholesterol is necessary in the brain for myelination, and formation of synapses and dendrites5-7. As much as cholesterol is needed for a healthy central nervous system, too much cholesterol can also cause detrimental neural effects, for example oxidative stress8,9, mitochondrial dysfunction10, inflammation9,11,12, and accumulation of a-synuclein13. All of which can contribute to the neurotoxicity of Parkinson’s Disease. There are conflicting data regarding Parkinson’s disease and cholesterol. Some studies have shown
Parkinson’s patients have lower levels of lower density lipoprotein (LDL) cholesterol and plasma cholesterol levels, however it is unclear whether decreased LDL levels declines after the onset of
Parkinson’s symptoms14,15. However, other clinical studies report that high levels of cholesterol increase the risk of developing Parkinson’s Disease.
Because statins are used to lower LDL cholesterol levels16, it is of interest to determine their relationship to Parkinson’s Disease. Evidence seems in favor of the use of statins leading to a decreased risk of Parkinson’s Disease, as studies show statins can limit neurotoxicity through suppression of pro-inflammatory cytokines17, limiting oxidative stress18, and prevent a-synuclein aggregation19, all of which can contribute to the loss of dopaminergic neurons in the nigrostriatal pathway. However, in the MPTP animal model of PD, the use of the drugs simvastatin and atorvastatin has been shown to damage nigrostriatal neurons20. In addition, clinical studies show limited association between the benefit of statins and Parkinson’s Disease and report statins may
119 increase the risk for Parkinson’s Disease21,22. If Parkinson’s patients have lower levels of LDL as reported by Huang14, Parkinson’s symptoms would be exacerbated if statins were taken due to further decreased cholesterol levels. Therefore, long term usage of statins may lead to an increase in risk for Parkinson’s Disease.
Another possible contributing factor of Parkinson’s Disease is the dysregulation of iron. Iron is necessary for many processes within the body such as oxygen transport in blood, DNA synthesis, and energy metabolism in the mitochondria23,24. Mutations within the HFE protein can result in the dysregulation of iron. The HFE protein is thought to contribute to iron regulation by binding to the iron transport protein, transferrin receptor 1 (TfR1)25,26. Mutations within this protein result in iron overload, hemochromatosis, typically in parenchymal cells that do not regulate iron27.
Two common HFE gene variants are C282Y and H63D27. C282Y is the most common mutation in hereditary hemochromatosis, however the H63D mutation is the most common mutation within the population28-30. The increased iron due to an HFE mutation can lead to increased reactive oxygen species31 and neuronal vulnerability32, both pathologies of neurodegenerative diseases. In addition, there have been studies linking HFE mutations, cholesterol and statin to neuronal dysfunction and diseases. Studies have shown C282Y mutation results in increased cholesterol within neuroblastoma cells, whereas H63D mutation has a decrease in neuroblastoma cells33. In addition, mice with H67D mutation, homolog to human H63D HFE, had decreased levels of brain cholesterol and memory impairment suggesting cholesterol levels may play a role in the contribution to Alzheimer’s Disease34. Furthermore, the H67D mice transfected with
Amyotrophic Lateral Sclerosis (ALS), SOD1(G93A) mutation, on a simvastatin diet resulted in accelerated disease progression and shorter survival35.
120 The aim of this study is to determine if a cholesterol diet or exposure to statins when coupled with the HFE mutation may impact the consequences of exposure to paraquat, a neurotoxicant used a model for studying Parkinson’s disease. We hypothesize that paraquat treatment will induce neurotoxicity, with the effects exaggerated by cholesterol diet, but protected by statin treatment.
5.3 Materials and Methods
5.3.1 Animals
Three-month old male mice containing wildtype (WT) or H67D, homolog of the human H63D
HFE mutation, alleles were generated in a mixture of C57BL/6 and 129/Sv mice. Genotype of mice were confirmed through tail clips in a process described in the previous chapter.
Mice were individually housed and given ad libitum access to rodent chow pellets and water. All mice were housed within an animal facility located at The Pennsylvania State University College of Medicine. The following procedures were approved by the Pennsylvania State University
College of Medicine Institutional Animal Care and Use Committee, protocol 46457, and conducted in accordance to the guidelines of the NIH Guide for the Care and Use of Laboratory
Animals.
5.3.2 Experimental Conditions
Wildtype or H67D mice (N=72) were divided into experimental treatment groups. Experimental groups consisted of mice receiving either saline (control) or 10 mg/kg paraquat dichloride hydrate
(Sigma Aldrich; St. Louis, MO) intraperitoneal injections once a week for three weeks, beginning at week 4. In addition, the mice were divided into three different diet groups (N=6/group).
Animals were fed either standard, control, rodent chow; or standard chow that was supplemented with 2% cholesterol or 6 mg of simvastatin per kg of chow (Harlan Tekland, Indianapolis, IN).
121 The animals continued with the same diet until the end of the study. Each week the animals were weighed, in addition to the amount of food consumed was recorded. Of note, mice on the control diet were also part of the experimental group in Chapter 4. The studies in Chapters 4 and this one in Chapter 4 were performed together.
5.3.3 Evaluation of Motor Function
Rotarod test was used to assess motor function. The mice were initially trained how to use the rotarod apparatus (Columbus Instruments; Columbus, OH) prior to any behavioral tests. On week two, the mice were trained on the rotarod assessment. The first behavioral test occurred on week three prior to paraquat or saline injection. The final behavioral test was at the end of week 6, following the last of the experimental treatments. Overall, the mice were placed on the rotarod for
180 seconds at 15 rpm. During the course of 180 seconds, the number of falls were recorded.
Motor performance was assessed by subtracting the number of falls from the second behavioral test from the first.
5.3.4 Statistical Analysis
Statistical analyses for behavior data and histologically sections were analyzed using the
Graphpad Prism (La Jolla, CA), in which the results were presented as mean ± SD. Statistical comparisons were completed by either a ungrouped t-test or a one-way ANOVA. A p-value of
≤0.05 was considered statistically significant.
5.4 Results
This study was carried out over the course of seven weeks, summarized in Figure 5-1. The study began by the mice being placed on one of three different rodent chow diets; control, simvastatin, or cholesterol. The animals were on the same diet throughout the seven-week study. Each week
122 the body weight of the animals was taken, in addition to the weight of the rodent chow to determine how much was consumed. In addition to examining dietary effects, the mice were further divided into treatment group receiving intraperitoneal injections of paraquat or saline.
These injections occurred once a week beginning at week three, for three weeks. Motor function and disease progression due to genotype, treatment, or diet was assessed by rotarod apparatus.
The mice were initially trained on rotarod a week two, and behavior tests were recorded at week three prior to the initial injection and week six following the last treatment injection.
123 Figure 5-1 Behavior test 2 (morning) & remove food in the evening
Control Behavior training Food & body weight the day before injection Diet Statin Behavior test 1 (the day Cholesterol before injection) Food & body weight
Food & body Food & body Food & body Harvest weight weight weight P P tissue Paraquat
0 1 2 3 4 5 6 7 Time (weeks)
Figure 5-1: Schematic Timeline of Experimental Design. The study was carried out for seven weeks, in which the mice were initially divided into three different diet groups receiving control, simvastatin, or cholesterol diet. This diet was continued throughout the seven weeks in which the amount of food consumed was recorded each week and body weight as well. Behavior training began at week two, with assessment recorded on weeks three and six. Lastly, experimental injections of paraquat or saline were divided between mice and given once a week for three weeks beginning on week 3 prior to the first behavior test.
124 5.4.1 Comparison of Diet Composition
Prior to behavioral analysis, the components of the diet were analyzed to determine if there were significant differences within the protein, carbohydrate and fat composition. There were no significant differences of macromolecules between the control, cholesterol, and simvastatin diets
(Figure 5-2).
125 Figure 5-2
% by weight % kcal from
Control Diet Protein 16.2 21.1 Carbohydrate 52.8 68.6 Fat 3.5 10.4 Kcal/g 3.1 Simvastatin Diet Protein 16.2 21.1 Carbohydrate 52.8 68.6 Fat 3.5 10.4 Kcal/g 3.1 Cholesterol Diet Protein 15.9 21.1 Carbohydrate 51.7 68.6 Fat 3.5 10.4 Kcal/g 3.0
Figure 5-2: Comparison of Diet Composition. Control, simvastatin, and cholesterol diets were compared to determine if there were any fundamental differences in protein, carbohydrate, and fat composition in terms of percent weight and percent kilocalories. There were no significant differences found between the different diets. Data represent the mean ± SD and are compared to all diets for statistical significance using 1-way ANOVA.
126 5.4.2 Changes in Body Weight and Mean Food Intake
In addition, we looked at how the different diets effected body weight and food intake. However, there was no significant differences within the change of body weight and mean food intake across the different diet (Figure 5-3).
127 Figure 5-3
Change in Body Weight (g) Mean Food Intake (kcal/kg)
Control Diet H67D Saline 0.8 ± 1.8 29.4 ± 2.4 Wt Saline -0.7 ± 2.9 27.5 ± 1.6 H67D PQ -1.4 ± 1.7 29.7 ± 4.4 Wt PQ 0.4 ± 1.3 32.5 ± 5.0 Simvastatin Diet H67D Saline 1.8 ± 4.5 33.0 ± 7.8 Wt Saline 1.6 ± 0.3 31.9 ± 0.6 H67D PQ -0.8 ± 4.9 31.2 ± 4.4 Wt PQ 1.6 ± 1.3 35.1 ± 8.6 Cholesterol Diet H67D Saline 0.3 ± 2.6 31.2 ± 7.5 Wt Saline -0.2 ± 2.6 32.7 ± 8.4 H67D PQ 0.8 ± 1.7 34.5 ± 3.7 Wt PQ 1.6 ± 1.3 31.2 ± 3.3
Figure 5-3: Experimental Group Comparison of Body Weight and Food Intake. Differences in body weights between experimental groups are presented as change in body weight. In addition, the average food take between the groups are compared as kcal of diet chow. There were no significant differences found between body weight or food intake. Data represent the mean ± SD and are compared to all diets for statistical significance using 1-way ANOVA.
128 5.4.3 Behavioral Differences between Genotype, Treatment, and Diet Groups
There are twelve experimental groups broken down into genotype, treatment, and diet. When comparing the behavior data of these ten groups, there was only a significant difference between two of the groups. Wildtype mice that received paraquat injections and a cholesterol diet had an increased number of falls compared to the wildtype mice on the control diet with saline injections
(Figure 5-4; p<0.01).
129 Figure 5-4 ** 10
5
0 Number of Falls -5
-10
Wt Saline-Control H67D Saline-Control Wt Paraquat-Control Wt Saline-Simvastatin Wt Saline-Cholesterol H767D Paraquat-Contol H67D Saline-Simvastatin Wt Paraquat-SimvastatinH67D Saline-Cholesterol Wt Paraquat-Cholesterol H67D Paraquat-Simvistatin H67D Paraquat-Cholesterol
Figure 5-4: Motor Behavioral Differences between Genotype, Treatment, and Diet Groups.
Motor function was assessed using a rotarod apparatus. The number of falls from second behavior test, following three injections, was subtracted from the number of falls from the first behavior test, prior to injections. An increase in the number of falls indicates a decrease in motor performance. Wildtype paraquat injected mice on a cholesterol diet had a significant number of falls compared to the wildtype saline injected mice on a control diet. There were no significant differences among the other groups. Data represent the mean ± SD and are compared to all treatment groups for statistical significance using 1-way ANOVA. **p<0.01 (n=5 or 6 per group).
130 5.4.4 Behavioral Differences Between Diet and Treatment, and Treatment and Genotype
The behavioral data was broken down into several other groups to determine the effect of the variables independently. Behavioral differences were examined between diet and treatment groups. H67D and wildtype mice were group together based upon diet and treatment. Overall, there were statistically significant differences between mice fed a control diet and received saline injections and mice receiving control diet and paraquat injections, cholesterol diet and saline injections, cholesterol diet and paraquat injections, and saline diet and paraquat injections (Figure
5-5; p<0.05 and p<0.01). Behavioral differences between genotype and treatment groups were also analyzed. The wildtype paraquat injected mice had a greater number of falls compared to the wildtype saline injected mice regardless of exposure to cholesterol in the diet or statin treatment
(Figure 5-6; p<0.05). There were no significant differences between the H67D mice.
131 Figure 5-5 Diet and Treatment 10 * * ** 5 *
0 Number of Falls -5
-10
Statin + Saline Control + Saline Statin + Paraquat Control + Paraquat Cholesterol + Saline Cholesterol + Paraquat
Figure 5-5: Effect of Diet and Treatment on Behavioral Motor Function. Mice were grouped based upon diet (control, cholesterol, or simvastatin) and injection they received. The number of falls from second behavior test, following three injections, was subtracted from the number of falls from the first behavior test, prior to injections. An increase in the number of falls indicates a decrease in motor performance. Animals on control diet with paraquat injections, cholesterol diet and saline injections, cholesterol diet and paraquat injections, or statin diet and paraquat injections had a greater number of falls compared to animals on control diet. Data represent the mean ± SD and are compared to all treatment groups for statistical significance using 1-way
ANOVA. *p<0.05 (n=11 or 12 per group).
132 Figure 5-6
10 *
5
0 Number of Falls -5
-10
Wt PQ H67D PQ Wt Saline H67D Saline
Figure 5-6: Effect of Treatment and Genotype on Behavioral Motor Function. Mice were grouped based upon genotype and injection they received. The number of falls from second behavior test, following three injections, was subtracted from the number of falls from the first behavior test, prior to injections. An increase in the number of falls indicates a decrease in motor performance. The wildtype paraquat injected mice, independent of diet, had an increase in number of falls compared to their saline counterpart. Whereas, there was no significant difference in the number of falls between the H67D treatment groups. Data represent the mean ± SD and are compared to all treatment groups for statistical significance using 1-way ANOVA. *p<0.05 (n=17 or 18 per group).
133 5.5 Discussion
The results of this study show that wildtype mice injected with paraquat had decreased motor function when fed a cholesterol diet. There was no effect of cholesterol and paraquat in the HFE mutant mice. Exposure to statins had no effect on the motor performance subsequent to paraquat injection on either genotype. Thus there is a genotype response to paraquat and cholesterol diet.
One goal of this study was to look at the induced paraquat effects of a cholesterol diet and statin treatment. Paraquat injections were associated with decreased motor function regardless of diet or exposure to statins compared to animals receiving only saline injections. Moreover, animals on cholesterol diet and saline injections had an increase in the number of falls compared to the control animals, indicating that cholesterol affected behavioral motor function even without paraquat injections. Indeed, in this group that was fed cholesterol, the paraquat injections did not have an additive effect. However, these groups contained both wildtype and H67D genotype mice. Therefore, we then looked at the effect of genotype on treatment. There was a decrease in motor function in the wildtype paraquat injected mice compared to the wildtype saline injected mice. There is no paraquat effect on the H67D. These results are similar to what was discussed in
Chapter 4 in which diet or statin exposure was not manipulated.
Our results showed there was a diet and genotype effect when the experimental groups were condensed. This lead us to then examine the effect of all three variables together. Of which the only significant finding was decreased motor function in wildtype mice on a cholesterol diet when injected with paraquat, compared to saline injected mice on a control diet. This would correspond with the theory that increased cholesterol may contribute to the onset of Parkinson’s
Disease, most likely due to increased neurotoxicity from oxidative stress and accumulation of a- synuclein. The H67D genotype did not exhibit significant changes in motor function in response
134 to paraquat when on a cholesterol or statin diet, indicating the genotype may be neuroprotective.
It is interesting to note that H67D mice have decreased levels of cholesterol within the brain34.
Because the brain synthesizes its own cholesterol4, it is not likely that the cholesterol diet resulted in an increase in brain cholesterol that led to the protective response to the paraquat injections.
Studies have shown increased systemic cholesterol can result in accumulation of LDL cholesterol in macrophages within arterial wall36. The macrophages release pro-inflammatory cytokines that are able to cross the blood brain barrier to exert oxidative stress upon the central nervous system37,38. Likewise, the absence of a statin effect, even with the decreased cholesterol in the brains of the H67D mice suggest statins are not effective in acute models of neurotoxicity34. The data are consistent with the concept that increased cholesterol may contribute to the onset of
Parkinson’s Disease, most likely due to increased neurotoxicity from oxidative stress and accumulation of a-synuclein, but the cholesterol effect is genotype specific.
In summary, when separating genotype, treatment, and diet, the combination of wildtype mice injected with paraquat and on a cholesterol diet, were the only animals exhibiting impaired motor functions. Therefore, the H67D genotype is still neuroprotective against paraquat even on a cholesterol or statin diet. The only time there was an effect on motor function with cholesterol or statin diet was when genotypes were combined. Indicating paraquat does induce neurotoxic effects, however, it is limited in the presence of the H67D genotype.
5.6 Acknowledgements
The work is supported in part by the Pennsylvania Department of Health Tobacco CURE Funds and the Penn State Neuroscience institute.
Thank you to Beth Neely for helping with the behavior study.
135 5.7 References
1. Hirsch EC. Biochemistry of Parkinson's disease with special reference to the
dopaminergic systems. Molecular neurobiology. 1994;9(1-3):135-142.
2. Polymeropoulos MH, Lavedan C, Leroy E, et al. Mutation in the alpha-synuclein gene
identified in families with Parkinson's disease. Science. 1997;276(5321):2045-2047.
3. Sayre LM, Perry G, Smith MA. Oxidative Stress and Neurotoxicity. Chemical Research
in Toxicology. 2008;21(1):172-188.
4. Zhang J, Liu Q. Cholesterol metabolism and homeostasis in the brain. Protein & Cell.
2015;6(4):254-264.
5. Dietschy JM, Turley SD. Thematic review series: Brain Lipids. Cholesterol metabolism
in the central nervous system during early development and in the mature animal. Journal
of Lipid Research. 2004;45(8):1375-1397.
6. Mauch DH, Nagler K, Schumacher S, et al. CNS synaptogenesis promoted by glia-
derived cholesterol. Science. 2001;294(5545):1354-1357.
7. Moutinho M, Nunes MJ, Correia JC, et al. Neuronal cholesterol metabolism increases
dendritic outgrowth and synaptic markers via a concerted action of GGTase-I and Trk.
Scientific Reports. 2016;6:30928.
8. Aytan N, Jung T, Tamtürk F, Grune T, Kartal-Özer N. Oxidative stress related changes in
the brain of hypercholesterolemic rabbits. Biofactors. 2008;33(3):225-236.
9. Thirumangalakudi L, Prakasam A, Zhang R, et al. High cholesterol-induced
neuroinflammation and amyloid precursor protein processing correlate with loss of
working memory in mice. Journal of neurochemistry. 2008;106(1):475-485.
10. Balboa E, Castro J, Pinochet M-J, et al. MLN64 induces mitochondrial dysfunction
associated with increased mitochondrial cholesterol content. Redox Biology.
2017;12:274-284.
136 11. Ullrich C, Pirchl M, Humpel C. Hypercholesterolemia in rats impairs the cholinergic
system and leads to memory deficits. Molecular and cellular neurosciences.
2010;45(4):408-417.
12. Xue Q-S, Sparks DL, Streit WJ. Microglial activation in the hippocampus of
hypercholesterolemic rabbits occurs independent of increased amyloid production.
Journal of Neuroinflammation. 2007;4(1):20.
13. Bosco DA, Fowler DM, Zhang Q, et al. Elevated levels of oxidized cholesterol
metabolites in Lewy body disease brains accelerate alpha-synuclein fibrilization. Nature
chemical biology. 2006;2(5):249-253.
14. Huang X, Chen H, Miller WC, et al. Lower low-density lipoprotein cholesterol levels are
associated with Parkinson's disease. Movement disorders : official journal of the
Movement Disorder Society. 2007;22(3):377-381.
15. Gudala K, Bansal D, Muthyala H. Role of serum cholesterol in Parkinson's disease: a
meta-analysis of evidence. Journal of Parkinson's disease. 2013;3(3):363-370.
16. Bilheimer DW, Grundy SM, Brown MS, Goldstein JL. Mevinolin and colestipol
stimulate receptor-mediated clearance of low density lipoprotein from plasma in familial
hypercholesterolemia heterozygotes. Proceedings of the National Academy of Sciences of
the United States of America. 1983;80(13):4124-4128.
17. Pahan K, Sheikh FG, Namboodiri AM, Singh I. Lovastatin and phenylacetate inhibit the
induction of nitric oxide synthase and cytokines in rat primary astrocytes, microglia, and
macrophages. J Clin Invest. 1997;100(11):2671-2679.
18. Barone E, Cenini G, Di Domenico F, et al. Long-term high-dose atorvastatin decreases
brain oxidative and nitrosative stress in a preclinical model of Alzheimer disease: a novel
mechanism of action. Pharmacological research. 2011;63(3):172-180.
137 19. Bar-On P, Crews L, Koob AO, et al. Statins reduce neuronal α-synuclein aggregation in
in vitro models of Parkinson's disease. Journal of neurochemistry. 2008;105(5):1656.
20. Kreisler A, Gele P, Wiart JF, Lhermitte M, Destee A, Bordet R. Lipid-lowering drugs in
the MPTP mouse model of Parkinson's disease: fenofibrate has a neuroprotective effect,
whereas bezafibrate and HMG-CoA reductase inhibitors do not. Brain Res.
2007;1135(1):77-84.
21. Huang X, Alonso A, Guo X, et al. Statins, plasma cholesterol and risk of Parkinson’s
disease: a prospective study. Movement disorders : official journal of the Movement
Disorder Society. 2015;30(4):552-559.
22. Liu G, Sterling NW, Kong L, et al. Statins may facilitate Parkinson's disease: Insight
gained from a large, national claims database. Movement disorders : official journal of
the Movement Disorder Society. 2017;32(6):913-917.
23. Wang J, Pantopoulos K. Regulation of cellular iron metabolism. The Biochemical
journal. 2011;434(3):365-381.
24. Andrews NC. Disorders of Iron Metabolism. New England Journal of Medicine.
1999;341(26):1986-1995.
25. Lebrón JA, Bennett MJ, Vaughn DE, et al. Crystal Structure of the Hemochromatosis
Protein HFE and Characterization of Its Interaction with Transferrin Receptor. Cell.
1998;93(1):111-123.
26. Bennett MJ, Lebron JA, Bjorkman PJ. Crystal structure of the hereditary
haemochromatosis protein HFE complexed with transferrin receptor. Nature.
2000;403(6765):46-53.
27. Feder JN, Gnirke A, Thomas W, et al. A novel MHC class I-like gene is mutated in
patients with hereditary haemochromatosis. Nat Genet. 1996;13(4):399-408.
138 28. Steinberg KK, Cogswell ME, Chang JC, et al. Prevalence of C282Y and H63D Mutations
in the Hemochromatosis (HFE) Gene in the United States. JAMA. 2001;285(17):2216-
2222.
29. Burt MJ, George PM, Upton JD, et al. The significance of haemochromatosis gene
mutations in the general population: implications for screening. Gut. 1998;43(6):830-836.
30. Marshall DS, Linfert DR, Tsongalis GJ. Prevalence of the C282Y and H63D
polymorphisms in a multi-ethnic control population. International journal of molecular
medicine. 1999;4(4):389-393.
31. Pulliam JF, Jennings CD, Kryscio RJ, et al. Association of HFE mutations with
neurodegeneration and oxidative stress in Alzheimer's disease and correlation with
APOE. American journal of medical genetics. Part B, Neuropsychiatric genetics : the
official publication of the International Society of Psychiatric Genetics. 2003;119b(1):48-
53.
32. Liu Y, Lee SY, Neely E, et al. Mutant HFE H63D protein is associated with prolonged
endoplasmic reticulum stress and increased neuronal vulnerability. The Journal of
biological chemistry. 2011;286(15):13161-13170.
33. Ali-Rahmani F, Huang MA, Schengrund CL, Connor JR, Lee SY. C282Y-HFE Gene
Variant Affects Cholesterol Metabolism in Human Neuroblastoma Cells. PLoS ONE.
2014;9(2):e88724.
34. Ali-Rahmani F, Grigson PS, Lee S, Neely E, Connor JR, Schengrund CL. H63D
mutation in hemochromatosis alters cholesterol metabolism and induces memory
impairment. Neurobiol Aging. 2014;35(6):1511.e1511-1512.
35. Su XW, Nandar W, Neely EB, Simmons Z, Connor JR. Statins accelerate disease
progression and shorten survival in SOD1(G93A) mice. Muscle & nerve.
2016;54(2):284-291.
139 36. Schwenke DC, Carew TE. Initiation of atherosclerotic lesions in cholesterol-fed rabbits.
I. Focal increases in arterial LDL concentration precede development of fatty streak
lesions. Arteriosclerosis (Dallas, Tex.). 1989;9(6):895-907.
37. Hansson GK, Robertson AK, Soderberg-Naucler C. Inflammation and atherosclerosis.
Annual review of pathology. 2006;1:297-329.
38. Yarlagadda A, Alfson E, Clayton AH. The Blood Brain Barrier and the Role of
Cytokines in Neuropsychiatry. Psychiatry (Edgmont). 2009;6(11):18-22.
140
Chapter 6
Underlying Mechanisms and Translational Implications of the H67D Mutation; Future Directions
6.1 Summary of Main Findings
As reviewed in chapter 1, iron is an essential element required for many essential physiological process, such as DNA synthesis, oxygen transport in the blood and energy metabolism within the mitochondria. Therefore, iron is tightly regulated and macrophages are a critical cell in maintaining iron homeostasis. The HFE gene is also involved in the regulation of iron homeostasis, and therefore mutations within the HFE gene can contribute to dysregulation of iron and generation of reactive oxygen species and free radicals. The HFE gene variant, H63D, has been found to alter iron handling within macrophages and is thought to influence the development of neurodegenerative diseases. Therefore, the main goal of this thesis is to study the implications of the H67D (mouse homolog of H63D) gene variant in macrophages in addition to the relationship between macrophage function and paraquat toxicity in a Parkinson’s Disease model.
Chapter 3 demonstrated the effect of H67D on macrophage phenotype. Through the use of bone marrow macrophages extracted from wildtype and H67D mice, the WT macrophages had an increase in proliferation and H67D macrophages had increased levels of L-ferritin upon iron exposure. In addition, the H67D macrophages had a slower migration rate and increased phagocytosis activity, in addition to several increased pro-inflammatory cytokines. Together, these results indicate there are genotype differences in iron handling and inflammatory immune response, in which the H67D alters macrophage function. Therefore, we examined role of H67D
141 in the context of Parkinson’s Disease, as peripheral and microglia mediated inflammation has been thought to contribute to the development and progression of neurodegenerative diseases1,2.
A Parkinson’s Disease-like model was induced through the neurotoxin, paraquat, which selectively targets the degeneration of dopaminergic neurons within the nigrostriatal pathway3.
This model is used in Chapter 4 to determine how the H67D genotype effected the neurodegeneration induced by paraquat. I found that the H67D genotype was neuroprotective against paraquat-induced dopaminergic neuron damage. This neuroprotection in H67D mice may stem from increased levels of L-ferritin storage within the substantia nigra, as free iron can induce oxidative stress and cell death.
The final chapter examined the effect of cholesterol and statins on H67D using the paraquat neurotoxic model, as cholesterol and statins has been linked to increased neurotoxicity in
Parkinson’s Disease. Of which the WT mice injected were paraquat were the only animals that exhibited changes in motor function, specifically when on a cholesterol diet. There were no changes in motor functions of the paraquat injected H67D mice on a cholesterol or statin diet.
Indicating, the statins may not have effect the neurotoxicity of paraquat and the H67D mice were protective against the toxic effects of cholesterol.
These findings suggest the H67D genotype exhibits a pro-inflammatory response in macrophages due to iron handling, in addition to neuroprotection in a Parkinsonism model. Together, these results warrant further study into the role of HFE gene variant effects on macrophages and other neurodegenerative diseases. Implications and future perspectives from these conclusions are discussed in the final chapter below.
142 6.2. A Little Stress Exerted by H67D can be Neuroprotective.
A major finding of this dissertation is the neuroprotective role of the H67D HFE genotype under paraquat-induced stress. H67D has shown to have increased levels of iron and oxidative stress within neuroblastoma cells, in addition to accelerating disease progression in an ALS model, therefore suggesting the H67D genotype would cause exacerbated effects of a neurotoxin4-7.
Despite the results of previous studies, the H67D mice exhibited a neuroprotective effect in the presence of paraquat, which may have been induced by external stressors due to the genotype. A key effect of the H67D mutation is iron accumulation, which is known to produce oxidative stress.
Too much or too little of iron within in the body is associated with negative physiological consequences. As iron transitions between its two states, Fe2+ and Fe3+, through the Fenton reaction it results in the generation of reactive oxygen species (ROS) and free radicals, see equation8 below.
(1) Fe2+ + H2O2 → Fe3+ + HO• + OH−
(2) Fe3+ + H2O2 → Fe2+ + HOO• + H+
Subsequent ROS and free radical production can lead to damage of cells by damaging their
DNA, protein, and lipid content9. Therefore, iron needs to be tightly regulated to avoid these detrimental effects. Importantly, mutations within the HFE gene have shown increased total body iron, including the brain10,11. Therefore, the effects of increased iron within the brain can result in oxidative stress and cell damage, in addition to progression of neurodegenerative diseases 12.
Corresponding to these theories, previous research has shown H67D mutation transfected SH-
SY5Y neuroblastoma cell line resulted in increased levels of iron, in addition to oxidative and endoplasmic reticulum stress as evident by increased levels of caspase-3 and lipid peroxidation4,5.
Furthermore, a double transgenic mouse consisting of the SOD1 (G93A) mutation and H67D
143 gene variant in a model of Amyotrophic Lateral Sclerosis (ALS), had an accelerated disease progression in the ALS mouse model7. These results led to the hypothesis that H67D in a paraquat-induced model of Parkinson’s Disease would result in increased oxidative damage and disease progression. Surprisingly, our work showed the H67D gene variant was neuroprotective against the paraquat toxicity (Chapter 4).
There are several mechanisms that may account for the neuroprotective role of H67D. primary candidate is the anti-oxidant NrF2 signaling pathway, which is thought to protect against the toxicity of paraquat13. Here, NrF2 is upregulated in the presence of oxidative stress through the
Keap1 (Kelch like ECH Associated Protein 1)-NrF2 pathway. Under normal physiological conditions, NrF2 is degraded when Keap1 binds to NrF2. However, in the presence of oxidative stress Keap1 dissociates from NrF2 allowing NrF2 to translocate to the nucleus where it binds to antioxidant response element (ARE). Activation of NrF2-ARE pathways results in the expression of genes encoding antioxidant enzymes such as, heme oxygenase-1 (HMOX1), glutathione S transferases (GSTs), and NAD(P)H quinone oxidoreductase 1 (NQO1)14-16.
Within the midbrain of Parkinson’s Disease patients, it is believed that the antioxidant defense mechanisms of the dopamine neurons are limited due to decreased levels of glutathione and peroxidases. However, within the remaining dopamine neurons of these patients, there was NrF2 within the nuclei, suggesting that NrF2 was protective of the remaining cells17. Transitioning into a Parkinson model induced by paraquat, PC12 cells (which have similar neurochemical and signal processes as dopamine neurons) treated with paraquat were found to have increased levels of
NrF2 and activated NrF2-ARE pathway protecting the cells against the neurotoxicity induced by paraquat13.
144 NrF2 also can minimize the effects of iron-induced oxidative stress independent to the canonical
NrF2-ARE pathway. Here, NrF2 can sequester free iron by increasing transcription of the two ferritin proteins, L-ferritin and H-ferritin18. Storage of iron within ferritin complex prevents the oxidation of iron by the Fenton equation. In support of this, I demonstrate that H67D mice had an increased level of L-ferritin compared to the wildtype (Chapter 4). In addition, patients with
H63D also had increased levels of brain L-ferritin19. Therefore, it would be expected to see an increase in L-ferritin within paraquat-treated H67D mice. However, this effect was not observed, as paraquat-treated H67D mice had reduced L-ferritin within the substantia nigra as compared to saline-treated control (Figure 4-6B). Therefore, if there were an increase in NrF2 in H67D mice it was independent of an increase in L-ferritin, and instead dependent upon oxidative stress.
Oxidative stress may have occurred within the H67D mice injected with saline, as the number of tyrosine hydroxylase (TH) positive neurons was approximately 70% less than with wildtype saline injected mice (Figure 4-5). When paraquat was introduced within the H67D mice there was not a significant decrease in TH neurons, indicating the neurons were protected from further damage. Thus it is possible that chronic oxidative stress caused the initial degeneration of TH neurons, and in return upregulated NrF2 to protect them from further damage by paraquat, by
NrF2 translocating to the nuclei of the dopaminergic neurons. These findings support the theory that NrF2 may be upregulated early on in H67D to protect the animals from additional stressors.
However, the initial stressors resulting in the damage of dopaminergic neurons within the H67D mice remains unclear. Such stressors may be due to an increase in pro-inflammatory cytokines released by macrophages, which will be discussed.
To determine the cause of decreased TH neurons within the H67D saline injected mice, we can look at Chapter 3 on how the H67D genotype affects macrophage phenotype. Macrophages can trigger a variety of physiological responses through their secretions, such as activating a pro-
145 inflammatory immune response, which can affect disease progression through peripheral inflammation1. The H67D macrophages have altered iron handling in response to iron loading
(increased L-ferritin; Figure 3-2A), differences in immune response (decreased migration and increased phagocytosis; Figure 3-5 and Figure 3-6), and differences in pro-inflammatory cytokines (increased chemoattractant proteins and decreased inflammatory proteins when stimulated with LPS; Figure 3-7). The connection between HFE mutant macrophages and disease progression may be triggered by their secretion of monocyte chemoattractant protein-1 (MCP-1).
MCP-1 is a chemoattractant protein involved in the recruitment and activation of additional macrophages and other monocytes usually in response to inflammation20. In addition, MCP-1 is involved in neuronal loss, as MCP-1 knock out mice have increased neuronal cell density compared to wildtype mice21. To further support this hypothesis, human neuroblastoma SH-
SY5Y cells transfected with H63D mutation have increased levels of MCP-122. In addition, H67D macrophages secrete increased levels of MCP-123 and H63D patients have increased serum MCP-
1 levels24. Thus, the increased levels of MCP-1 reported in H63D/H67D studies may be the reason for the decreased levels of TH neurons seen within the substantia nigra of the H67D mice, prior to paraquat injections. From these results, it led us to the hypothesis that the H63D/H67D macrophages secrete MCP-1 which passes through the blood brain barrier (BBB) to cause oxidative stress and neuronal loss within the substantia nigra. From the increased cell death and oxidative stress due to MCP-1, NrF2 would be upregulated and translocated to the nuclei of TH neurons to protect them from additional stressors such as paraquat. Upon the injection of paraquat, peripheral macrophages would be activated and secrete additional chemokines. This theory is supported by work in Chapter 3 demonstrating H67D macrophages have increased levels of the chemokines, JE and MIP-2, when stimulated by LPS. However, the pro- inflammatory effects of the chemokines did not create additional damage to the substantia nigra as seen in the WT mice, likely due to the neuroprotective effect of NrF2. To test this theory, we
146 can use H67D and WT macrophage-conditioned media on H67D and WT dopaminergic neurons to examine the toxic or neuroprotective effects of the macrophage secretions. Table 6-1 describes the setup and hypothesized results.
To determine if NrF2 is indeed the mechanism of H67D-mediated neuroprotection, subsequent studies should be performed investigating NrF2 protein and mRNA levels within brain tissue homogenates, in addition to NrF2 targets such as HMOX1, GST, and NQ01. Furthermore, expression of NrF2 within the substantia nigra will be assessed to determine if NrF2 is translocated to the nuclei of dopamine neurons for neuroprotection. Moreover, NrF2 may have translocate to other cells, such as glia, where they can promote secretion anti-inflammatory cytokines to protect against toxins and cell death.
In order to combat the oxidative stress exerted on dopaminergic neurons, NrF2 can be used as a neuroprotective therapy. To increase levels of NrF2, you can disrupt the interaction between
Keap1-NrF2 or upregulate NrF2 through NrF2 activators. Currently there is only one clinical
NrF2 activator which is used to treat multiple sclerosis25. In Parkinson’s Disease, NrF2 therapy would be most beneficial if used at the initiation of degeneration of dopaminergic neurons.
However, diagnosis occurs when clinical symptoms appear, in which 60-80% of dopaminergic neurons have already been degenerated26. Therefore, early detection would be required for optimal benefits of a NfR2 therapeutic.
147 Table 6-1
Macrophage Conditioned Media
H67D Wildtype
Increased cell death No change in cell death H67D
Increased cell death No change in cell death Dopaminergic Neurons Dopaminergic Wildtype Table 6-1: Experimental Design of H67D Macrophage Conditioned Media on Dopaminergic
Neurons. Conditioned media from H67D and wildtype macrophages will be added to wildtype and transfected H67D dopaminergic neurons. The results will indicate whether the secretions from the macrophages are protective or toxic to dopaminergic neurons. Conditioned media from
H67D macrophages are predicted to cause cell death in both H67D and wildtype macrophages, with further increased cell death in H67D macrophages due to an increase in oxidative stress that has been found within H63D SH-SY5Y neuroblastoma cell line.4 There should not be cell death observed in dopaminergic neurons cultured with conditioned media from wildtype macrophages.
However, there may be a slight increase in cell death within the H67D neurons due to the oxidative stress previously stated.
148 6.3 The Varying Role of H67D in an ALS Mouse Model
The neuroprotective role of H67D shown within the paraquat model of Parkinson’s Disease
(Chapter 4), led to the thought of the role of H67D in other neurodegenerative diseases. Our lab previously investigated the role of H67D genotype in a double transgenic ALS mouse model of
SOD1(G93A) and H67D mutations. However, the H67D genotype did not show a neuroprotective role, as animals with both SOD1 and H67D mutations resulted in an accelerated disease progression compared to SOD1 and WT mice7. The differences in the role of H67D poses the question of how the H67D genotype can be both neuroprotective and neurotoxic in disease progression. In order to answer this question, the two systems must be compared.
Within the ALS model, the mice with the accelerated disease progression were bred to contain two mutations, SOD1(G93A) and H67D, both that are known to elicit oxidative stress6,27. These mutations could have evoked oxidative stress from birth. Whereas, in the Parkinsonism model the second “stressor”, paraquat was given three months after birth and only for an acute time period.
In addition, these are also different disease models. The ALS model targets all motor neurons from the spinal cord to the brain whereas paraquat is specific to targeting dopaminergic neurons in the nigrostriatal pathway. In regard to motor function of the two mouse models, the paraquat injected H67D mice did not have an impaired motor function, which may have been attributed to the lack of significant difference of TH neurons between H67D saline and paraquat injected mice.
The SOD1/H67D mice did have significantly impaired motor function compared to the SOD1 and WT mice. However, there was no significant difference between the number of motor neurons found in the SOD1 mice and SOD1/H67D mice7, indicating that the H67D mice may have had protective effect on motor neurons within ALS model, despite the increased levels of oxidative stress found. The selective sensitivity of dopaminergic neurons in HFE models versus motor neurons may be due to the presence of dopamine, as recently reported, upregulation of the
149 dopamine transporter on dopamine neurons results in the selective death of dopaminergic neurons28. Future studies should look at the TH neurons in the nigrostriatal pathway in the
SOD1/H67D mice, as clinical studies have shown diminished TH neurons in ALS patients29. The
SOD1/H67D mice should maintain the same number of TH neurons as the SOD1 indicating the
H67D genotype is neuroprotective against further neuron lost as supported by the motor neuron, and H67D paraquat data. Furthermore, theories in subsection 6.2 suggest that the loss of H67D neurons was due to early stress.
Lastly, the neurotoxicity found within the double transgenic mice could be due to an increase number of copies of the SOD1 mutation, or the diet of the mice. When the study of ALS progression was repeated, the SOD1/H67D mice did not result in accelerated disease progression
(results not published), which may have been due to a different number of SOD1 copies or more importantly the differences in diet. When the mice were on a diet with higher cholesterol and carbohydrate levels, the SOD1/H67D had an increased survival compared to the SOD1 mice, indicating H67D and cholesterol together are neuroprotective against ALS. These results were similar to my findings in Chapter 5, in which H67D mice on a cholesterol diet, injected with paraquat, did not have a significant difference in behavior. Whereas WT mice fed a cholesterol diet had an increased number of falls when injected with paraquat. This data suggests that diet and cholesterol levels may influence the impact of HFE on neurodegeneration.
150 6.4 Macrophage Therapy in Parkinson’s Disease.
There is currently no cure for Parkinson’s Disease, instead any treatment options extend to improving motor function due to the loss of dopaminergic neurons. However, the therapy options only treat the symptoms of Parkinson’s Disease and do not prevent further loss of dopaminergic neurons. Therefore, therapy options need to include the rescue of the remaining dopamine neurons, and protect them from additional cell loss. To prevent further cell loss, therapeutic options should target the underlying cause of disease progression. The mechanism regarding dopaminergic cell death is unknown, but thought to be contributed to inflammation and oxidative damage30. Importantly, macrophages and microglia have been speculated to be the primary cells involved in inflammation found in Parkinson’s Disease, and thus, may provide promising targets for therapeutics2,31,32.
Macrophages/Microglia have the remarkable ability to adapt to their surrounding and activate into killer (M1) or helper (M2) subtype of cells. In terms of Parkinson’s Disease, macrophages/microglia are typically thought to be in M1 activated states. M1 macrophages/microglia are the classic activation state associated with a killing function through the secretion of pro-inflammatory cytokines, such as TNFa and IL-6. Induction of M1 macrophages/microglia occurs in the presence of interferon gamma (IFNg) and LPS activating
JAK/STAT pathway and toll-like receptors (TLR) which activates a variety of pro-inflammatory cytokines33. In comparison, M2 macrophages are a broader set of macrophages involved in healing, and induced by IL-4 and IL-10, resulting in the secretion of anti-inflammatory cytokines33. Therefore, it may be beneficial attenuate the response of M1 macrophages/microglia and stimulate them into M2 activation state to promote neuroprotection. To inhibit the response of the M1 activation state, the downstream pathway, JAK/STAT pathway and TLRs, can be blocked to prevent the release of pro-inflammatory cytokines34. M1 responses can also be
151 inhibited by converting M1 macrophages and microglia into M2 state, through introducing IL-4 and IL-1035,36. These proteins can be packaged into nanovesicles that are able to cross the blood brain barrier to activate microglia within the brain37.
6.5 The Potential Role of C282Y in Macrophage Phenotype and Disease Progression.
This thesis has examined the HFE gene variant, H63D, or the mouse homolog H67D; however, there is another HFE mutation, C282Y that has yet to be looked at in the content of these chapters. The C282Y mutation is not as common in the population but it is more prevalent within patients with hereditary hemochromatosis38-40. Patients with the C282Y mutation present with increased levels of serum iron and serum transferrin saturation, whereas H63D patients have mild to no iron overload symptoms41. Therefore, we speculate whether the C282Y mutation would reproduce the results of H63D.
The questions of this thesis addressed the role of H63D genotype on macrophage phenotype and disease progression in a paraquat induced Parkinsonism model, in which the H67D genotype resulted in impaired iron handling and immune response in macrophages and neuroprotection against paraquat injected mice. As C282Y genotype is associated with increased body iron, but iron poor macrophages, the C282Y macrophages should have decreased levels of the iron uptake protein transferrin receptor (TfR) or increased levels of the iron export protein, ferroportin. In addition, iron can be regulated by hepcidin through its ability to bind to ferroportin, thus preventing iron export42. Therefore, C282Y macrophages may have decreased levels of hepcidin which would result in decreased levels of cellular iron. The dysregulation of iron by C282Y macrophages would also result in an impaired immune function, as indicated by the increase in oxidative stress exhibited by SH-SY5Y neuroblastoma cells transfected with C282Y genotype4.
Due to an increase in oxidative stress, the C282Y mutations would have increased pro-
152 inflammatory cytokine secretions leading to increased neuronal vulnerability. Therefore, C282Y mice injected with paraquat should result in a decrease in TH neurons and motor function. In comparison to WT mice, the C282Y mice may have an increased motor impairment due to the pro-inflammatory cytokines released by macrophages prior to paraquat injections. It is not known if the C282Y mice exhibit a decrease in motor neurons as shown in H67D saline injected mice.
However, SH-SY5Y neuroblastoma cells transfected with C282Y mutation had decreased levels of MCP-1 secretions compared to WT cells22 and no association between C282Y patients and serum MCP-1 levels24. Therefore, it would be predicted that the C282Y mice would not have a
TH neuronal loss prior to paraquat injections, and NrF2 would not be upregulated to evoke a neuroprotective response as theorized in H67D mouse model. However, this is not to say that the
C282Y mice would not be protected against the toxic effects of paraquat, as clinical studies have found the C282Y mutation to be protective against the development of Parkinson’s Disease43.
6.6 Does the H67D Genotype Exert Long Term Neuroprotection in Paraquat Mouse Model?
Chapters 4 and 5 utilized an acute model in which mice were given paraquat once a week for three weeks. Following, the mice were transcardially perfused for analysis by immunohistochemistry. All though this mouse model is effective in determining the relationship of genotype on paraquat induced neurotoxicity, this was only an acute model and lacks the long term neurodegeneration effects associated with Parkinson’s Disease. Thus, a chronic model would examine the long term effects of paraquat, and may a provide a more physiological system.
Previous reports have shown acute Parkinsonism models having short term results in which the behavior and neurological deficits are reversed44. With a chronic model, continual exposure to paraquat would result in a deficit of motor function in the H67D mice compared to H67D saline injected mice. Paraquat would exert additional stress on the dopaminergic neurons and on the macrophages within the peripheral system as well. Continual stress on macrophages would result
153 in increased pro-inflammatory cytokines secretions, therefore oxidative stress. A further increase in oxidative stress, in addition to the hypothesized stress of MCP-1, can result in further upregulation of NrF2, therefore limiting cell death. In contrast, an increase in oxidative stress could also result in decreased NrF2 levels, as NrF2 would not be able to offset the increase in oxidative stress. Previous studies have shown increased levels of NrF2 following oxidative stress, however continuation of oxidative stress then resulted in reduced NrF2 levels. Thus, NrF2 may provide neuroprotection for short term oxidative stress but not in long term, chronic oxidative states. Therefore, if NrF2 cannot protect from the increased oxidative stress, the H67D mice would result in a rapid degeneration of motor function as they already had approximately 70% less TH neurons compared to WT saline injected mice (Figure 4-5).
6.7 The Role of Cholesterol and Statins in H67D Macrophages
Previous sections of this chapter talked about the role of macrophages in systemic inflammation that may contribute to the pathogenesis or neuroprotection seen in WT and H67D paraquat mouse models. However, what was not elucidated from Chapter 5, is the role the macrophages played in neuroprotection in the model of paraquat exposure coupled with statin treatment.
Wildtype mice on a cholesterol diet receiving paraquat injections were the only group from
Chapter 5 that had an impaired motor function (Figure 5-4). Increased levels of cholesterol are thought to contribute to the pathogenesis of Parkinson’s disease through increased inflammation, accumulation of a-synuclein, and oxidative stress. In terms of our Parkinsonism model, cholesterol should be accumulating within macrophages, in which increased macrophage cholesterol results in secretion of pro-inflammatory cytokines45-47. However, the cholesterol diet had no effect on motor function within the H67D mice, indicating there was a neuroprotective function of the genotype. Previous sections within this chapter have elucidated on the reasons
154 behind this, however, none discussed the role of the genotype and cholesterol. Despite studies showing macrophages accumulate cholesterol, resulting in an inflammatory response, there are studies showing macrophages can resist the buildup of cholesterol, limiting inflammatory response48. The H67D macrophages may secrete an extrinsic mediator that prevents the accumulation of cholesterol or they simply do not take up cholesterol, therefore limiting the secretion of pro-inflammatory cytokines. Previous studies report the H67D mutant’s inability to uptake cholesterol, as H63D transfected SH-SY5Y neuroblastoma cells had 50% less cholesterol compared to WT SH-SY5Y cells. In addition, decreased brain cholesterol levels were observed within H67D mice49. The lack of cholesterol uptake may be mediated by altered expression of cholesterol regulatory proteins due to increased oxidative stress.
There were no significant differences in motor function between WT and H67D genotype when the mice were on a simvastatin diet. This corresponds to majority of literature supporting statin use and decreased risk for developing Parkinson’s Disease. However, there are a few clinical studies that believe statin use is associated with development of Parkinson’s Disease or unmasks
Parkinson’s Disease50,51. A previous study reported simvastatin promotes pro-inflammation when induced by LPS52. This could support clinical studies showing statin use could lead to the progression or unmasking of Parkinson’s Disease. However, the question is why we did not see these effects in our model. With the H67D mice, it could be due to the protection of the genotype but we did not see an effect of statin on WT mice as well. The LPS study on simvastatin was completed in-vitro, however in an in-vivo model, if the statins were to induce an inflammatory response when mice were treated with statin, it could be due to the pro-inflammatory cytokines unable to cross the blood brain barrier. A previous study showed pro-inflammatory cytokines were able to cross the BBB due to cholesterol increasing the leakiness of the BBB through increased vascular inflammation53. To verify the effect found with simvastatin and LPS, H67D
155 and WT macrophages should be exposed directly to paraquat and simvastatin. The results should predict a pro-inflammatory response from the WT macrophages, but a limited to no inflammatory response in the H67D macrophages due to their protective nature. Furthermore, due to the decrease brain cholesterol reported within H67D mice49, it would be expected that the use of statins would further decrease cholesterol levels, therefore increasing inflammation and unmasking the paraquat effect. However, this result was not observed and may be attributed to the delivery of statins. Simvastatin was chosen as it is able to pass through the BBB, however simvastatin was supplemented into the diet. Therefore, after digestion the amount of statin remaining may not have been significant to cross the BBB and increase the oxidative stress on the remaining TH neurons within the H67D mice.
6.8 Bone Marrow Macrophages vs. Peritoneal Macrophages
Peritoneal (PEMs) and bone marrow derived macrophages (BMMs) are the most common in vitro models used in macrophages studies, of which only bone marrow derived macrophages were used throughout the course of this dissertation. The use of BMMs was due to the high yield, proliferation, and phagocytosis capabilities. In contrast, BMMs are less differentiated and immature, whereas, PEMs are more mature and functionally stable54. In a quick study of L- ferritin with PEMs, the H67D PEMs had increased levels of L-ferritin compared to wildtype
PEMs (Appendix), which is the result expected in Chapter 3. This difference indicates that there are extracellular factors contributing to the loss of L-ferritin within PEMs and not BMMs.
However, within the isolation of peritoneal macrophages, there are also a number of B and T cells, therefore not giving a pure culture. In addition, to isolate a sufficient number of peritoneal macrophages to use in a study, thioglycollate is injected into the peritoneal cavity. Despite the fact this increases macrophage yield, it may also change the physiology of peritoneal macrophages, as it is increases macrophage migration. Therefore, the use of bone marrow derived
156 macrophages in a naïve state may more accurately represent the physiology as peritoneal macrophages are activated upon isolation.
6.9 Circuitry of the Substantia Nigra
Literature reports of Parkinson’s Disease mainly focuses on the motor paradigm affecting patients. However, in addition to the motor impairments, Parkinson’s patients are also affected cognitively, in which they report slow thinking or the inability to concentrate. The basal ganglia, which contains the substantia nigra, controls both motor and cognitive function through interactions with the cortex and cerebellum. The cognitive decline in Parkinson’s patients may not be only due to decreased dopamine levels, but loss of norepinephrine and acetylcholine receptors may contribute as well. As chapters 4 and 5 only focused on testing motor function, continued studies should look at cognitive function as well. To determine how the H67D mutation also effects cognitive function in the Parkinsonism paraquat model, the mice could undergo an object recognition test and water maze. The results of this test should show that the
H67D mice have cognitive impairment, of which would be decreased upon paraquat injection.
Previous studies with H67D mice have already reported a cognitive decline compared to wildtype49, in addition paraquat treated mice report increase cognitive deficits as well55.
6.10 Conclusions
Prior to the research conducted in this thesis, it was reasoned that the H67D genotype would accelerate disease progression due to increased oxidative stress. However, the results of my thesis indicate the impact of the H67D genotype can extend from disrupting immune function within macrophages to neuroprotection in a Parkinsonism model, opening a new field to understand its role in disease progression. The impaired macrophages affect the role of H67D in disease models such as Parkinsonism, possibly by secreting pro-inflammatory cytokines such as MCP-1. The
157 mechanism behind this hypothesis needs to be elucidated but the result of the peripheral inflammation exerted by the macrophages would create oxidative stress early in the H67D mice, and therefore may result in the upregulation of NrF2 to protect the mice from further damage. In conclusion, the H67D genotype can result in oxidative stress through impacting macrophage phenotype and contributing to disease progression. However, the stress exerted by the H67D gene variant makes it more resilient to additional stressors that can led to protection from additional disease consequences.
158 6.11 References
1. Träger U, Tabrizi SJ. Peripheral inflammation in neurodegeneration. J Mol Med.
2013;91(6):673-681.
2. Qian L, Flood PM. Microglial cells and Parkinson's disease. Immunologic research.
2008;41(3):155-164.
3. Brooks AI, Chadwick CA, Gelbard HA, Cory-Slechta DA, Federoff HJ. Paraquat elicited
neurobehavioral syndrome caused by dopaminergic neuron loss. Brain Research.
1999;823(1–2):1-10.
4. Lee SY, Patton SM, Henderson RJ, Connor JR. Consequences of expressing mutants of
the hemochromatosis gene (HFE) into a human neuronal cell line lacking endogenous
HFE. The FASEB Journal. 2007;21(2):564-576.
5. Liu Y, Lee SY, Neely E, et al. Mutant HFE H63D protein is associated with prolonged
endoplasmic reticulum stress and increased neuronal vulnerability. The Journal of
biological chemistry. 2011;286(15):13161-13170.
6. Nandar W, Neely EB, Unger E, Connor JR. A mutation in the HFE gene is associated
with altered brain iron profiles and increased oxidative stress in mice. Biochimica et
Biophysica Acta (BBA) - Molecular Basis of Disease. 2013;1832(6):729-741.
7. Nandar W, Neely EB, Simmons Z, Connor JR. H63D HFE genotype accelerates disease
progression in animal models of amyotrophic lateral sclerosis. Biochimica et Biophysica
Acta (BBA) - Molecular Basis of Disease. 2014;1842(12, Part A):2413-2426.
8. Fenton H. Oxidation of tartaric acid in presence of iron. J Chem Soc. 1894;65:899-903
9. Pham-Huy LA, He H, Pham-Huy C. Free Radicals, Antioxidants in Disease and Health.
International Journal of Biomedical Science : IJBS. 2008;4(2):89-96.
159 10. Nielsen JE, Jensen LN, Krabbe K. Hereditary haemochromatosis: a case of iron
accumulation in the basal ganglia associated with a parkinsonian syndrome. Journal of
neurology, neurosurgery, and psychiatry. 1995;59(3):318-321.
11. Berg D, Hoggenmuller U, Hofmann E, et al. The basal ganglia in haemochromatosis.
Neuroradiology. 2000;42(1):9-13.
12. Batista-Nascimento L, Pimentel C, Andrade Menezes R, Rodrigues-Pousada C. Iron and
Neurodegeneration: From Cellular Homeostasis to Disease. Oxidative Medicine and
Cellular Longevity. 2012;2012:8.
13. Izumi Y, Yamamoto N, Matsushima S, et al. Compensatory role of the Nrf2–ARE
pathway against paraquat toxicity: Relevance of 26S proteasome activity. Journal of
Pharmacological Sciences. 2015;129(3):150-159.
14. Kensler TW, Wakabayashi N, Biswal S. Cell survival responses to environmental stresses
via the Keap1-Nrf2-ARE pathway. Annual review of pharmacology and toxicology.
2007;47:89-116.
15. Reichard JF, Motz GT, Puga A. Heme oxygenase-1 induction by NRF2 requires
inactivation of the transcriptional repressor BACH1. Nucleic Acids Research.
2007;35(21):7074-7086.
16. Itoh K, Chiba T, Takahashi S, et al. An Nrf2/small Maf heterodimer mediates the
induction of phase II detoxifying enzyme genes through antioxidant response elements.
Biochem Biophys Res Commun. 1997;236(2):313-322.
17. Ramsey CP, Glass CA, Montgomery MB, et al. Expression of Nrf2 in Neurodegenerative
Diseases. Journal of neuropathology and experimental neurology. 2007;66(1):75-85.
18. Huang B-W, Ray PD, Iwasaki K, Tsuji Y. Transcriptional regulation of the human
ferritin gene by coordinated regulation of Nrf2 and protein arginine methyltransferases
PRMT1 and PRMT4. The FASEB Journal. 2013;27(9):3763-3774.
160 19. Bartzokis G, Lu PH, Tishler TA, et al. Prevalent Iron Metabolism Gene Variants
Associated with Increased Brain Ferritin Iron in Healthy Older Men. Journal of
Alzheimer's disease : JAD. 2010;20(1):333-341.
20. Deshmane SL, Kremlev S, Amini S, Sawaya BE. Monocyte Chemoattractant Protein-1
(MCP-1): An Overview. Journal of Interferon & Cytokine Research. 2009;29(6):313-
326.
21. Sawyer AJ, Tian W, Saucier-Sawyer JK, et al. The effect of inflammatory cell-derived
MCP-1 loss on neuronal survival during chronic neuroinflammation. Biomaterials.
2014;35(25):6698-6706.
22. Mitchell RM, Lee SY, Randazzo WT, Simmons Z, Connor JR. Influence of HFE variants
and cellular iron on monocyte chemoattractant protein-1. Journal of Neuroinflammation.
2009;6:6-6.
23. Weston C, Hund W, Nixon, A, et al. Host H67D Genotype Affects Tumor Growth in
Mouse Melanoma. Journal of Cancer Science & Therapy. 2015;7:216-223.
24. Lawless MW, White M, Mankan AK, O'Dwyer MJ, Norris S. Elevated MCP-1 serum
levels are associated with the H63D mutation and not the C282Y mutation in hereditary
hemochromatosis. Tissue antigens. 2007;70(4):294-300.
25. Bomprezzi R. Dimethyl fumarate in the treatment of relapsing–remitting multiple
sclerosis: an overview. Therapeutic Advances in Neurological Disorders. 2015;8(1):20-
30.
26. Marsden CD. Parkinson's disease. The Lancet. 1990;335(8695):948-949.
27. Barber SC, Mead RJ, Shaw PJ. Oxidative stress in ALS: A mechanism of
neurodegeneration and a therapeutic target. Biochimica et Biophysica Acta (BBA) -
Molecular Basis of Disease. 2006;1762(11):1051-1067.
161 28. Leyva-Illades D, Chen P, Zogzas CE, et al. SLC30A10 Is a Cell Surface-Localized
Manganese Efflux Transporter, and Parkinsonism-Causing Mutations Block Its
Intracellular Trafficking and Efflux Activity. The Journal of Neuroscience.
2014;34(42):14079-14095.
29. Kato S, Oda M, Tanabe H. Diminution of dopaminergic neurons in the substantia nigra of
sporadic amyotrophic lateral sclerosis. Neuropathology and Applied Neurobiology.
1993;19(4):300-304.
30. Tansey MG, Goldberg MS. Neuroinflammation in Parkinson’s disease: its role in
neuronal death and implications for therapeutic intervention. Neurobiology of disease.
2010;37(3):510-518.
31. Long-Smith CM, Sullivan AM, Nolan YM. The influence of microglia on the
pathogenesis of Parkinson's disease. Prog Neurobiol. 2009;89(3):277-287.
32. Funk N, Wieghofer P, Grimm S, et al. Characterization of peripheral hematopoietic stem
cells and monocytes in Parkinson's disease. Movement disorders : official journal of the
Movement Disorder Society. 2013;28(3):392-395.
33. Mosser DM, Edwards JP. Exploring the full spectrum of macrophage activation. Nature
reviews Immunology. 2008;8(12):958-969.
34. Liu Y, Holdbrooks AT, De Sarno P, et al. Therapeutic Efficacy of Suppressing the
JAK/STAT Pathway in Multiple Models of EAE(). Journal of immunology (Baltimore,
Md : 1950). 2014;192(1):59-72.
35. Joniec-Maciejak I, Ciesielska A, Wawer A, et al. The influence of AAV2-mediated gene
transfer of human IL-10 on neurodegeneration and immune response in a murine model
of Parkinson's disease. Pharmacological Reports. 2014;66(4):660-669.
162 36. Schwenkgrub J, Joniec-Maciejak I, Sznejder-Pacholek A, et al. Effect of human
interleukin-10 on the expression of nitric oxide synthases in the MPTP-based model of
Parkinson's disease. Pharmacological reports : PR. 2013;65(1):44-49.
37. Yang T, Martin P, Fogarty B, et al. Exosome Delivered Anticancer Drugs Across the
Blood-Brain Barrier for Brain Cancer Therapy in Danio Rerio. Pharmaceutical research.
2015;32(6):2003-2014.
38. Feder JN, Gnirke A, Thomas W, et al. A novel MHC class I-like gene is mutated in
patients with hereditary haemochromatosis. Nat Genet. 1996;13(4):399-408.
39. Simon M, Bourel M, Fauchet R, Genetet B. Association of HLA-A3 and HLA-B14
antigens with idiopathic haemochromatosis. Gut. 1976;17(5):332-334.
40. Merryweather-Clarke AT, Pointon JJ, Shearman JD, Robson KJ. Global prevalence of
putative haemochromatosis mutations. Journal of medical genetics. 1997;34(4):275-278.
41. Adams PC, Reboussin DM, Barton JC, et al. Hemochromatosis and Iron-Overload
Screening in a Racially Diverse Population. New England Journal of Medicine.
2005;352(17):1769-1778.
42. Nemeth E, Tuttle MS, Powelson J, et al. Hepcidin Regulates Cellular Iron Efflux by
Binding to Ferroportin and Inducing Its Internalization. Science. 2004;306(5704):2090-
2093.
43. Buchanan DD, Silburn PA, Chalk JB, Le Couteur DG, Mellick GD. The Cys282Tyr
polymorphism in the HFE gene in Australian Parkinson's disease patients. Neuroscience
Letters. 2002;327(2):91-94.
44. Lau Y-S. Progressive Neurodegeneration in the Chronic MPTP/Probenecid Model of
Parkinson?s Disease. Parkinson's Disease: CRC Press; 2004.
163 45. Schwenke DC, Carew TE. Initiation of atherosclerotic lesions in cholesterol-fed rabbits.
I. Focal increases in arterial LDL concentration precede development of fatty streak
lesions. Arteriosclerosis (Dallas, Tex). 1989;9(6):895-907.
46. Hansson GK, Robertson AK, Soderberg-Naucler C. Inflammation and atherosclerosis.
Annual review of pathology. 2006;1:297-329.
47. Yarlagadda A, Alfson E, Clayton AH. The Blood Brain Barrier and the Role of
Cytokines in Neuropsychiatry. Psychiatry (Edgmont). 2009;6(11):18-22.
48. Spann Nathanael J, Garmire Lana X, McDonald Jeffrey G, et al. Regulated Accumulation
of Desmosterol Integrates Macrophage Lipid Metabolism and Inflammatory Responses.
Cell. 2012;151(1):138-152.
49. Ali-Rahmani F, Grigson PS, Lee S, Neely E, Connor JR, Schengrund CL. H63D
mutation in hemochromatosis alters cholesterol metabolism and induces memory
impairment. Neurobiol Aging. 2014;35(6):1511.e1511-1512.
50. Huang X, Alonso A, Guo X, et al. Statins, plasma cholesterol and risk of Parkinson’s
disease: a prospective study. Movement disorders : official journal of the Movement
Disorder Society. 2015;30(4):552-559.
51. Liu G, Sterling NW, Kong L, et al. Statins may facilitate Parkinson's disease: Insight
gained from a large, national claims database. Movement disorders : official journal of
the Movement Disorder Society. 2017;32(6):913-917.
52. Matsumoto M, Einhaus D, Gold ES, Aderem A. Simvastatin Augments
Lipopolysaccharide-Induced Proinflammatory Responses in Macrophages by Differential
Regulation of the c-Fos and c-Jun Transcription Factors. The Journal of Immunology.
2004;172(12):7377.
164 53. Sparks DL, Kuo YM, Roher A, Martin T, Lukas RJ. Alterations of Alzheimer's disease in
the cholesterol-fed rabbit, including vascular inflammation. Preliminary observations.
Ann N Y Acad Sci. 2000;903:335-344.
54. Wang C, Yu X, Cao Q, et al. Characterization of murine macrophages from bone
marrow, spleen and peritoneum. BMC Immunology. 2013;14(1):6.
55. Chen L, Yoo SE, Na R, Liu Y, Ran Q. Cognitive impairment and increased Abeta levels
induced by paraquat exposure are attenuated by enhanced removal of mitochondrial
H(2)O(2). Neurobiol Aging. 2012;33(2):432.e415-426.
Appendix
Peritoneal Macrophage L-ferritin Concentration
Figure 7-1
DataData 1 1 500 500 H67DH67D **** 400 400 WildtypeWildtype
300 300
200 200 ** ** 100 100 Intracellular L-ferritin (ug ) (ug L-ferritin Intracellular Intracellular L-ferritin (ug ) (ug L-ferritin Intracellular 0 0
FAC FAC DFO DFO Neutral Control
Figure 7.1: L-ferritin Concentration within Peritoneal Macrophage Cellular Lysates.
Peritoneal macrophages were plated in 100 mm2 tissue culture dishes (8 x 106 cells/flask).
Following maturation, the cells were incubated with control, 300 uM FAC, or 300 uM DFO supplemented media for 24 hours. Following the media was replaced with control media for an additional 24 hours. After 24 hours, cells were harvested and lysed for analysis using an L-ferritin
ELISA. Absorbance of the cell lysates was measured on a SpectraMax 340PC plate reader. Data represent the mean±SEM from three independent experiments and are compared to all treatment groups for statistical significance using 2-way ANOVA. ** p<0.01, ****p<0.0001
VITA Anne Marie Nixon Education 2017 Doctor of Philosophy in Anatomy The Pennsylvania State University College of Medicine, Hershey, PA 2010 Bachelor of Science in Biobehavioral Health; Minor in Biology The Pennsylvania State University, University Park, PA
Teaching Experience 2017 Adjunct Professor at Shippensburg University, Shippensburg, PA 2016 Adjunct Professor at Shippensburg University, Shippensburg, PA Teaching Assistant, Human Gross Anatomy in Medical and Physician Assistant School Curriculum; Penn State Hershey College of Medicine, Hershey, PA Cadaver Prosector for the Ultrasound-Guided Cadaver Course in Regional Anesthesia; Penn State Hershey College of Medicine 2015 Teaching Assistant, Human Gross Anatomy in Medical and Physician Assistant School Curriculum; Penn State Hershey College of Medicine, Hershey, PA Cadaver Prosector for the Ultrasound-Guided Cadaver Course in Regional Anesthesia; Penn State Hershey College of Medicine 2014 Teaching Assistant, Human Gross Anatomy and Neuroanatomy in Medical and Physician Assistant School Curriculum; Penn State Hershey College of Medicine, Hershey, PA Lecturer for Neuroanatomy Residents; Penn State Hershey College of Medicine, Hershey, PA Cadaver Prosector for the Ultrasound-Guided Cadaver Course in Regional Anesthesia; Penn State Hershey College of Medicine 2013 Teaching Assistant, Human Gross Anatomy and Neuroanatomy in Medical School Curriculum; Penn State Hershey College of Medicine, Hershey, PA Lecturer for Neuroanatomy Residents; Penn State Hershey College of Medicine, Hershey, PA
Honors and Awards 2017 Shippensburg University Academic Mentorship Award International BioIron Society Travel Award Penn State Graduate Student Travel Award 2016 American Association of Anatomists (AAA) Travel Award
Publications Nixon AM, Meadowcroft MD, Neely E, Purnell CJ, Nandar W, Huang X, Connor JR. The Neuroprotective Role of the HFE Gene Variant H67D in a Paraquat Mouse Model. Submitted Nixon AM, Neely E, Simpson IA, Connor JR. The Role of HFE Genotype in Macrophage Phenotype. Submitted Nixon AM, Connor JR (2017) Does HFE Genotype Impact Macrophage Phenotype in Disease Process and Therapeutic Response? Biometals in Neurodegenerative Diseases: Mechanisms and Therapeutics. Adhikary SD, El-Boghdadly K, Nasralah Z, Sarwani N, Nixon AM, Chin KJ. A radiologic and anatomic assessment of injectate spread following transmuscular quadratus lumborum block in cadavers. Anaesthesia. 2017;72(1):73-79. Schonberg DL, Miller TE, Wu Q, Flavahan WA, Das NK, Hale JS, Hubert CG, Mack SC, Jarrar AM, Karl RT, Rosager AM, Nixon AM, Tesar PJ, Hamerlik P, Kristensen BW, Horbinski C, Connor JR, Fox PL, Lathia JD, Rich JN. (2015) Preferential Iron Trafficking Characterizes Glioblastoma Stem-like Cells. Cancer Cell 28(4): 441-455. Connor JR, Zhang X, Nixon AM, Webb B, Perno JR (2015) Comparative Evaluation of Nephrotoxicity and Management by Macrophages of Intravenous Pharmaceutical Iron Formulations. PLoS ONE 10(5) Weston C, Hund W, Nixon A, Neely E, Webb B, Alkhateeb A, Connor JR. (2015) Host H67D Genotype Affects Tumor Growth in Mouse Melanoma. J Cancer Sci Ther 7:216-223.