Zinc and the Zinc Transporter SLC39A10/ZIP10 are Required for Heme Synthesis in Developing Erythroid Progenitors A DISSERTATION SUBMITTED TO THE FACULTY OF THE UNIVERSITY OF MINNESOTA BY Juyoung Kim IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY Advisor: Dr. Moon-Suhn Ryu January 2022 Copyright © Juyoung Kim, 2022 All rights reserved. i Acknowledgment I would like to express my sincere appreciation to my advisor, Dr. Moon-Suhn Ryu, for his guidance and patience during my doctoral studies. His continuous support and motivation have encouraged me to pursue my graduate career. I also thank my committee members, Dr. Xiaoli Chen, Dr. Dan Gallaher, and Dr. Sarah Cusick, for their time, support, and advisement for this research. I would like to acknowledge Dr. Jaekwon Lee for his help in the quantitative metal analyses. I would like to extend my thanks to my lab colleagues, Emily Bengson and Cole Guggisberg, and to all faculty and staff in the department. Lastly, I thank my friends and family for their love and endless support. ii Abstract Zinc is an essential trace element for various biological processes in the body. Cellular zinc homeostasis is established via differential expressions of the cell-type- specific zinc transporters, ZnTs and ZIPs. Zinc deficiency in humans has been associated with anemia. Yet, molecular mechanisms by which zinc alters red blood cell development remain uncertain. The current studies elucidate the roles of cellular zinc during red blood cell development and identify a zinc transporter important in maintaining erythroid zinc homeostasis during terminal erythroid differentiation and zinc deficiency. G1E-ER4 and MEL cells were employed for the characterization of the roles of zinc and zinc transporter in terminal erythroid differentiation and heme biosynthesis. G1E-ER4 cells featured a 1.7-fold increase in total cellular zinc contents after 48-hour of differentiation. Acute zinc depletion by a membrane-permeable chelator TPEN resulted in a rapid loss in cell viability. Restriction of extracellular zinc supply using a membrane- impermeable zinc chelator DTPA impaired hemoglobinization of differentiating G1E- ER4 cells without a significant loss in cell viability. The decreased heme contents by DTPA were fully restored by replenishing equimolar zinc and were not due to changes in total cellular iron contents. Zinc-deficient G1E-ER4 cells differentiated with normal Alas2 transcript response but less Alad and alpha-globin mRNA abundance. Deficiency of the heme precursor metabolite, protoporphyrin, was observed in MEL cells experiencing heme deficiency due to restricted zinc supply. Among the 24 zinc transporter genes, Zip10 was identified as most responsive to cellular zinc deficiency in developing G1E-ER4 cells. Moreover, the upregulation of ZIP10 was found essential for iii adequate hemoglobin production when cellular zinc supply was restricted. Through screening of the gnomAD database, we have identified three missense ZIP10 variants specifically polymorphic among the African population. The presence of these polymorphisms was experimentally confirmed by genotyping genomic DNA from African individuals. The current studies characterize zinc as a nutrient essential to proper erythroid maturation and heme biosynthesis and identify a compensatory role of ZIP10 for erythroid zinc homeostasis under zinc restriction. The findings suggest that poor zinc status and ZIP10 mutations might serve as potential risk factors and are expected to be new therapeutic targets for erythrocyte-related disorders, including anemia. iv Table of Contents Acknowledgment ............................................................................................................................. i Abstract ........................................................................................................................................... ii Table of Contents .......................................................................................................................... iv List of Tables ................................................................................................................................ vii List of Figures .............................................................................................................................. viii List of Abbreviations .................................................................................................................... xi Chapter 1 . Introduction................................................................................................................ 1 Introduction ................................................................................................................................ 2 Zinc biology ................................................................................................................................ 3 Biochemical roles of zinc ......................................................................................................... 3 Zinc as an essential nutrient .................................................................................................... 4 Cellular zinc homeostasis ........................................................................................................ 6 Systemic zinc homeostasis ........................................................................................................ 8 Zinc in erythroid development and red cell metabolism ........................................................ 9 Zinc and anemia ....................................................................................................................... 9 Erythropoiesis ........................................................................................................................ 11 Erythropoietic gene regulations............................................................................................. 11 Zinc-containing metalloenzyme in mature red cell ................................................................ 12 Heme biosynthesis .................................................................................................................. 13 Zinc transporter expression in the erythron .......................................................................... 14 Metal ions in nutritional immunity ........................................................................................ 15 Hypoferremia and hypozincemia by inflammation ................................................................ 15 Zinc transporter variants in diseases ..................................................................................... 16 Metal transporter variants in immunity against endemic infections ..................................... 17 Hypothesis and study aims ...................................................................................................... 19 Chapter 2 . Zinc is required for heme synthesis in developing erythroid progenitors .......... 24 Introduction .............................................................................................................................. 25 Materials and Methods ............................................................................................................ 27 Maintenance of cells and induction for differentiation .......................................................... 27 Zinc treatments....................................................................................................................... 28 Major minerals and trace elements analyses ......................................................................... 28 v Total cell counts and cell viability ......................................................................................... 28 RNA isolation, Reverse Transcription, and quantitative RT-PCR ......................................... 28 Western analyses .................................................................................................................... 29 RNA-seq transcriptome analyses and data processing .......................................................... 30 Heme, ZnPP, and total PPIX assays ...................................................................................... 31 Statistics ................................................................................................................................. 31 Results ....................................................................................................................................... 34 Differentiation of G1E-ER4 cells ........................................................................................... 34 The cellular zinc pool expands during the terminal erythroid differentiation ....................... 34 Zinc is required for hemoglobinization .................................................................................. 35 Zinc deficiency does not change cellular iron contents but may affect the cellular distribution of iron. ................................................................................................................ 37 Transcriptome profile of zinc-deficient erythroid progenitors .............................................. 38 Differential responses of heme biosynthetic genes to zinc restriction ................................... 39 Protoporphyrin metabolism during zinc restriction. ............................................................. 40 Zinc deficiency does not globally impair GATA1-induced gene expression. ......................... 42 Discussion ................................................................................................................................. 65 Chapter 3 . Zinc transporter ZIP10 is required for heme biosynthesis and cell survival during red cell development ........................................................................................................ 71 Introduction .............................................................................................................................. 72 Materials and Methods ............................................................................................................ 73 Cell line and treatments ......................................................................................................... 73 Gene silencing by short interfering RNA (siRNA) ................................................................. 74 Major minerals and trace elements analyses ......................................................................... 74 Quantitative RT-PCR and Western blot ................................................................................. 74 Heme assays ........................................................................................................................... 74 Transcriptome and bioinformatic analyses ............................................................................ 74 Statistics ................................................................................................................................. 75 Results ....................................................................................................................................... 78 Zip10 is the most responsive gene to zinc restriction among 24 zinc transporters ............... 78 Zip10 is suppressed during terminal erythroid differentiation but is highly upregulated by zinc deficiency. ....................................................................................................................... 79 ZIP10 functions an essential role in hemoglobinization under zinc restriction during red cell development. .......................................................................................................................... 80 ZIP10 deficiency impairs zinc homeostasis during red cell development under zinc restriction. .............................................................................................................................. 82 vi ZIP10 depletion under zinc restriction is associated with ferroptosis cell death. ................. 82 Discussion ............................................................................................................................... 107 Chapter 4 . The identification of genetic variants of ZIP10 among the African population ..................................................................................................................................................... 114 Introduction ............................................................................................................................ 115 Materials and Methods .......................................................................................................... 116 SNP screening through the database ................................................................................... 116 Predicted topology of ZIP10 and the location of its mutations ........................................... 116 Whole blood and DNA sample preparation of the African population ................................ 116 Genotyping ........................................................................................................................... 117 Genotype-based whole blood assessment ............................................................................ 117 Results ..................................................................................................................................... 117 The three notable ZIP10 SNPs specific to the African population are identified through public databases................................................................................................................... 117 The topology of human ZIP10 and the location of its mutations in the evolutionary alignment of the gene ............................................................................................................................ 119 The ZIP10 SNPs are present in the African population ....................................................... 120 ZIP10 SNP specific to the African population is associated with hematological indices in the Ugandan children ................................................................................................................ 121 Discussion ............................................................................................................................... 135 Chapter 5 . Summary ............................................................................................................... 141 List of references ........................................................................................................................ 145 Appendices .................................................................................................................................. 161 vii List of Tables Table 2-1. Primers for SYBR Green-based qPCR analyses .......................................................... 33 Table 2-2. The total cellular contents of minerals in proliferating and differentiating erythroid progenitors. .................................................................................................................................... 44 Table 2-3. Differentially expressed genes (DEGs) of GATA1-target genes by zinc restriction and the prediction on GATA1 activity identified by IPA. .................................................................... 63 Table 3-1. ZnT primers for SYBR Green-based qPCR analyses. .................................................. 76 Table 3-2. Zip primers for SYBR Green-based qPCR analyses .................................................... 77 Table 3-3. Affected canonical pathways by ZIP10 depletion and zinc restriction during red cell development. ................................................................................................................................ 102 Table 3-4. Differential expression of ferroptosis-related genes by ZIP10 depletion and zinc restriction in developing G1E-ER4 cells. .................................................................................... 106 Table 4-1. Three missense genetic variants specific to the African population in the SLC39A10/ZIP10 gene (gnomAD database). .............................................................................. 122 Table 4-2. Three missense genetic variants specific to the African population in the SLC39A10/ZIP10 gene (1000 Genome Project). ......................................................................... 123 Table 4-3. Genotyping of ZIP10 SNP (rs13419724) in Ugandan children. ................................. 128 Table 4-4. Genotyping of ZIP10 SNPs in the African population. .............................................. 132 Table 4-5. Hematological indices of ZIP10 SNP (rs13419724) in Ugandan children ................. 134 viii List of Figures Figure 1-1. Location and direction of zinc transporters and a scheme of transcriptional regulation via MTF-1 in the nucleus. .............................................................................................................. 21 Figure 1-2. A scheme of terminal erythroid differentiation. .......................................................... 22 Figure 1-3. A scheme of heme biosynthesis in the cytosol and mitochondria. .............................. 23 Figure 2-1. Temporal patterns of transcripts during terminal erythroid differentiation. ................ 43 Figure 2-2. Cellular zinc restriction impairs hemoglobinization in developing G1E-ER4 cells. .. 45 Figure 2-3. Zinc restriction impairs heme production from the early stage of terminal erythroid differentiation in G1E-ER4 cells. .................................................................................................. 46 Figure 2-4. The total cell counts, cell viability, and live cell size are not affected by zinc restriction at 24 hours of G1E-ER4 cell differentiation but are reduced at 48 hours of differentiation. ................................................................................................................................ 47 Figure 2-5. Decreased heme contents by DTPA treatment are fully restored by adding back equimolar zinc. ............................................................................................................................... 48 Figure 2-6. Reduced cellular zinc contents by DTPA are fully replenished by adding back equimolar zinc. ............................................................................................................................... 49 Figure 2-7. Adequate zinc supply is essential for heme biosynthesis at the early stage of the terminal erythroid differentiation. .................................................................................................. 50 Figure 2-8. Restriction of zinc supply decreases alpha-globin mRNA and protein abundance in developing G1E-ER4 cells. ............................................................................................................ 51 Figure 2-9. Zinc restriction has no effect on total cellular iron contents during red cell development. .................................................................................................................................. 52 Figure 2-10. Cytosolic labile iron pool is not altered by DTPA treatment at 24-hour of differentiation but may be affected by DTPA at 48-hour of differentiation in G1E-ER4 cells. .... 53 Figure 2-11. Hierarchical cluster analysis of differential expressions by zinc restriction discriminates zinc-deficient developing G1E-ER4 cells from zinc-adequate proliferating and developing G1E-ER4 cells. ............................................................................................................ 54 Figure 2-12. Volcano plot of DEGs by zinc restriction in developing G1E-ER4 cells. ................ 55 Figure 2-13. ALAD is the most responsive to zinc restriction among heme biosynthetic genes. . 56 Figure 2-14. Zinc restriction does not impair Alas2 expression but represses Alad mRNA and protein expression. ......................................................................................................................... 57 ix Figure 2-15. The accumulation of total protoporphyrin IX is decreased by zinc restriction during erythroid terminal differentiation in G1E-ER4 cells. ..................................................................... 58 Figure 2-16. MEL-DS19 cells are suitable for analyzing the effects of zinc status on porphyrin metabolites. .................................................................................................................................... 59 Figure 2-17. MEL-DS19 cells treated with DMSO undergo hemoglobinization. ......................... 60 Figure 2-18. Hemoglobin-related genes are elevated in MEL-DS19 treated with DMSO. ........... 61 Figure 2-19. Zinc deficiency immediately diminishes zinc protoporphyrin levels in MEL-DS19 cells treated with DMSO. ............................................................................................................... 62 Figure 3-1. Zip10 is the most responsive gene to zinc restriction among 24 zinc transporters in developing erythroid progenitors. .................................................................................................. 84 Figure 3-2. TPEN treatment leads to significant cell loss in proliferating and developing G1E- ER4 cells. ....................................................................................................................................... 85 Figure 3-3. Transcript responses of zinc transporter genes to TPEN treatment in differentiating G1E-ER4 Cells. ............................................................................................................................. 86 Figure 3-4. Zip10 transcript immediately and directly responds to zinc deficiency. ..................... 87 Figure 3-5. Temporal patterns of transcripts of zinc transporters and metallothionein during terminal erythroid differentiation. .................................................................................................. 88 Figure 3-6. Zinc restriction immediately increases Zip10 mRNA levels during the terminal erythroid differentiation. ................................................................................................................ 89 Figure 3-7. Zinc restriction increases ZIP10 protein abundance during red cell development...... 90 Figure 3-8. Heme deficiency is caused by different factors in developing G1E-ER4 cells. .......... 91 Figure 3-9. Zip10 only responds to zinc restriction, not to heme deficiency caused by iron or protoporphyrin deficiency during red cell development. ............................................................... 92 Figure 3-10. Zip10 is responded to zinc restriction by DTPA and fully reversed by the addition of equimolar zinc. ............................................................................................................................... 93 Figure 3-11. Gene silencing approach using ZIP10 siRNA suppresses Zip10 mRNA abundance. ....................................................................................................................................................... 94 Figure 3-12. ZIP10 depletion alone did not produce changes in molecular indices of heme biosynthesis during red cell development. ..................................................................................... 95 Figure 3-13. Zip10 responds to zinc restriction by DTPA in a dose-dependent manner. .............. 96 Figure 3-14. ZIP10 is required under zinc restriction for heme production during red cell development. .................................................................................................................................. 97 x Figure 3-15. ZIP10 depletion results in the impairments of alpha-globin protein and transcript under zinc restriction during red cell development. ....................................................................... 98 Figure 3-16. ZIP10 depletion reduces cellular mineral contents under zinc restriction during red cell development. ........................................................................................................................... 99 Figure 3-17. ZIP10 depletion affects Mt1 expressions from the early terminal erythroid development under zinc restriction. ............................................................................................. 100 Figure 3-18. Venn diagrams of differentially expressed genes by zinc restriction and ZIP10 depletion during red cell development. ........................................................................................ 101 Figure 3-19. ZIP10 depletion reduces cell viability under mild zinc restriction during the terminal erythroid differentiation. .............................................................................................................. 105 Figure 4-1. The population-based distribution map of the three ZIP10 SNPs which are specific to the African population. ................................................................................................................ 124 Figure 4-2. A predicted topological model of human SCL39A10/ZIP10. .................................... 125 Figure 4-3. Evolutionary alignment of human, chimpanzee, mouse, and rat ZIP10 amino acid sequences with the variants of interest. ........................................................................................ 126 Figure 4-4. An allelic discrimination plot of rhAmp SNP Assays targeting SLC39A10 SNP, rs13419724 among the Ugandan children. .................................................................................. 127 Figure 4-5. An allelic discrimination plot of rhAmp SNP Assays targeting SLC39A10 SNP, rs13419724 among the African population. ................................................................................. 129 Figure 4-6. An allelic discrimination plot of rhAmp SNP Assays targeting SLC39A10 SNP, rs114992984 among the African population. ............................................................................... 130 Figure 4-7. An allelic discrimination plot of rhAmp SNP Assays targeting SLC39A10 SNP, rs76182486 among the African population. ................................................................................. 131 Figure 4-8. The implication of ZIP10 SNP (rs13419724) on blood levels of hemoglobin and zinc protoporphyrin in the Ugandan children ...................................................................................... 133 xi List of Abbreviations ABCA1 ATP-binding cassette subfamily A member 1 ABCB6 ATP Binding Cassette Subfamily B Member 6 AE Acrodermatitis Enteropathica ALAD δ-Aminolevulinic Acid Dehydratase ALAS2 5-Aminolevulinic Acid Synthase2 BME β-Mercaptoethanol BMSCs Bone Marrow Stromal Cells CA Carbonic Anhydrase CHO Chinese Hamster Ovary CO III Coproporphyrinogen III CPOX Coproporphyrinogen Oxidase CRP C-Reactive Protein DEGs Differentially Expressed Genes DFO Deferoxamine DMEM Dulbecco's Modified Eagle Medium DMSO Dimethylsulfoxide DTPA Diethylenetriamine Pentaacetate EKLF Erythroid Kruppel-Like Factor EPO Erythropoietin ER Endoplasmic Reticulum FANCD2 Fanconi Anemia Complementation Group D2 FC Fold Change FDFT1 Farnesyltransferase 1 Fe Iron FECH Ferrochelatase FPN Ferroportin GATA1 GATA-Binding factor 1 gnomAD Genome Aggregation Database h Hour H2az1 H2A.Z Variant Histone 1 HAMP Hepatic Antimicrobial Peptide Hba-a1/2 Hemoglobin alpha adult chain 1 and 2 Hbb-b1 Hemoglobin beta adult chain 1 HMB Hydroxymethylbilane HMBS Hydroxymethylbilane Synthase HMGCR 3-Hydroxy-3-Methylglutaryl-CoA Reductase xii HMOX1 Heme Oxygenase-1 HRI Heme-Regulated Inhibitor ICP-MS Inductively Coupled Plasma-Mass Spectrometry IDA Iron Deficiency Anemia IL-6 Interleukin 6 IMDM Iscove’s Modified Dulbecco’s Medium IPA Ingenuity Pathway Analysis IRP Iron Regulatory Protein 2 LOOH Lipid Reactive Oxygen Species LPS Lipopolysaccharide MEP Megakaryocyte-Erythroid Progenitor MFRN1 Mitoferrin-1 min Minutes MRE Metal Response Elements MT Metallothionein MTF-1 Metal-response element-binding Transcription Factor-1 n.s. Not Significant PAGE Polyacrylamide Gel Electrophoresis PBG Porphobilinogen PBS Phosphate Buffered Saline Pol II Polymerase II PPG Protoporphyrinogen PPIX Protoporphyrin IX PPOX Protoporphyrinogen Oxidase qPCR Quantitative Polymerase Chain Reaction RBL1 RB (retinoblastoma) transcriptional corepressor like 1 RIN RNA integrity number RNAi RNA interference ROS Reactive Oxygen Species RT Room Temperature Sat1 Spermidine/spermine N1-acetyltransferase 1 SDS Sodium Dodecyl Sulfate siCON Control siRNA-transfected cells siRNA short interfering RNA siZIP10 ZIP10 siRNA-transfected cells SLC Solute-Carrier gene SNP Single Nucleotide Polymorphism SOD Superoxide Dismutase SVM Support Vector Machines xiii TM Transmembrane Protein TPEN N,N,N′,N′-tetrakis(2-pyridylmethyl)-ethylenediamine TSPO Mitochondrial Translocator Protein URO Uroporphyrinogen UROD Uroporphyrinogen Decarboxylase UROS Uroporphyrinogen Synthase Yap1 Yes-related protein 1 ZIP Zrt-,Irt like Protein Zn Zinc ZnA Zinc-Adequate ZnD Zinc-Deficient ZnPP Zinc Protoporphyrin ZnT Zinc Transporter ZPP/H Zinc Protoporphyrin : Heme β-Est β-Estradiol δ-ALA δ-Aminolevulinic Acid 1 Chapter 1 . Introduction 2 Introduction In the human body, zinc serves as an essential nutrient involved in diverse biological processes, including immunity, growth, and development. Zinc deficiency is a public health problem with an estimated global prevalence of 31% (1), and has been associated with anemia risk and impaired red blood cell development. Plasma and serum zinc have been proposed as an indicator for anemia risk in humans (2–5). Tissue zinc acquisition of the bone marrow increased during the recovery from hemolytic anemia has been demonstrated in mice, suggesting the requirement of zinc for compensatory erythropoiesis (6). Moreover, zinc deficiency in rats can lead to impaired membrane integrity of circulating red blood cells, which may lead to higher risks of hemolytic anemia (7). Recently, at the cellular level, it has been demonstrated that intracellular erythroid zinc serves for cell survival at the early stage of terminal erythroid differentiation (8). Yet, the effects of cellular zinc status on hemoglobin production, a critical process of erythroid development and of the etiology of anemia, have not been demonstrated, and the molecular mechanisms by which erythroid cells maintain cellular zinc homeostasis remain poorly understood. This thesis presents the roles of zinc in erythroid cells and identifies a gene essential in the maintenance of adequate cellular zinc supply during mild zinc restriction. This chapter reviews the current understandings and relevant literature of the physiological roles of zinc, zinc transporters in cellular zinc homeostasis, erythroid development, and red blood cell disorders. Thereafter, the effects of cellular zinc restriction on red cell development, particularly hemoglobinization, the role and 3 regulation of a zinc importer protein essential for erythroid development, and SLC39A10/ZIP10 SNPs uniquely present in Africans will be introduced and discussed. The following thesis concludes with a summary of the overall findings and addresses their implications for future research and clinical and dietetic practices. Zinc biology Biochemical roles of zinc Zinc is an essential trace element that is involved in diverse biological processes in the body, including immunity, growth, and development (9). In cells, zinc functions as a structural, catalytic, and regulatory component of metalloproteins and signaling pathways (10). Erythroid Kruppel-like factor (EKLF) and GATA-Binding factor 1 (GATA1), the master transcriptional regulators of erythroid development, are exemplars of zinc-finger proteins that use zinc as a structural component (11,12). Another well- characterized metalloenzyme containing zinc as a structural component is the Cu/Zn superoxide dismutase (SOD). SOD1 with an oxidoreductase function serves to scavenge reactive oxygen species (ROS) and mitigate oxidative stress (13). Copper serves as the catalytic cofactor of SOD1 in this process. Metalloenzymes requiring zinc as a catalytic cofactor can be categorized into six classes; oxidoreductase, transferase, hydrolase, lyase, isomerase, and ligase (14). Among the zinc-dependent lyases, erythrocyte carbonic anhydrase (CA1) is where zinc mediates the conversion of water and carbon dioxide into bicarbonate (10,15). Human and mouse δ-aminolevulinic acid dehydratase (ALAD), a key enzyme mediating heme biosynthesis by catalyzing the conversion of δ- aminolevulinic acid (δ-ALA) to porphobilinogen (PBG), is another lyase that requires 4 zinc as a cofactor (10,16). Recent studies have revealed the functions of zinc as a signaling molecule of pathways mediating immune or apoptotic responses, which represent its regulatory roles (17). Moreover, zinc has been characterized as a regulatory component for the synthesis, storage, secretion, and stabilization of a hormone integral to glucose metabolism, insulin (18,19). Zinc as an essential nutrient Zinc deficiency in humans was initially documented in the early 1960s, yet remains a highly prevalent nutrition disorder with an estimated global prevalence as high as 31 % (20). The first case of zinc deficiency was reported in 1961 with the symptoms of dwarfism, hypogonadism, dermatitis, and iron deficiency anemia (IDA) (21,22). According to the original case report, the patient had undernourished meal habits, such as eating clay daily, wheat bread with a small amount of milk, and rarely consuming protein foods. As a type 2 nutrient, zinc is involved in and required for various molecular and biological processes throughout the human body and thus does not feature a specific deficiency disorder (23,24). The demand for zinc in the human body varies depending on age, lifecycle, and other health conditions. Early signs of acute zinc deficiency include anorexia, reduced taste and smell perception, as termed dysgeusia and dysosmia, respectively, and altered emotionalities (e.g., anxiety, irritability, and depression-like behavior) in humans (25–27). Acute zinc deficiency also may manifest as diarrhea, eye abnormalities, and severe dermatitis accompanied by alopecia and impaired wound healing (28,29). In children, chronic zinc deficiency causes impairment of physical and sexual development and increases the risk of mortality by infectious diseases, such as 5 pneumonia (30,31). In adulthood, zinc is essential in maintaining healthy reproductive and immunological systems (32,33). Additionally, zinc deficiency has been associated with anemia in human (34), and increased membrane osmotic fragility of erythrocytes by low zinc have been proposed as a potential etiology (35,36). Yet, there is a significant gap in knowledge of the molecular pathways and genetic components mediating the impact of zinc deficiency on red cell development and anemia. The current recommendations for zinc adequacy are intakes of zinc at 11 mg and 8 mg for adult men and women, respectively (37). The daily upper limit of zinc in adults is 40 mg and primarily concerns the potential effects of excess zinc on nausea, vomiting, lethargy, and fatigue (38). Zinc is in a variety of food sources such as shellfish, meat, and legumes (37). Yet, some plant foods, including legumes and whole grains, contain phytate, which can interfere with the absorption of zinc from its food sources by binding to the metal atom (39). A phytate:zinc molar ratio above 18 has been estimated to negatively affect the bioavailability of zinc from the diet (40). In America's diet, red meat and poultry are the major food sources for zinc intake (41). To date, supplementation of dietary zinc has been accepted as a major therapeutic strategy for zinc deficiency (42). However, the efficacy of zinc treatment remains still controversial. Although zinc toxicity rarely occurs when consumed from natural sources, overdose by supplements can lead to symptoms of nausea, diarrhea, and headaches (43). Pharmacological doses of zinc can interfere with copper absorption, and secondary sideroblastic anemias and leukopenia associated with zinc-induced copper deficiency have been reported (44). More recently, neurological impairments by copper deficiency 6 have been attributed to unintended intake of excessive zinc from the usage of denture cream products with high zinc contents (45,46). Cellular zinc homeostasis Cellular zinc homeostasis is established through coordinated regulations of import, distribution, storage, and efflux. Zinc transporters play a fundamental role in zinc mobilization across plasma and intracellular membranes. In contrast to iron, which has a central regulator for its systemic distribution, zinc requires cell-type-specific regulatory mechanisms due to its ubiquitous essentiality (10). The genes of zinc transporters are assigned to the solute-carrier (SLC) gene superfamily of membrane transport proteins (35). There are two families of zinc transporters; the ZnT, and ZIP (Zrt-,Irt like protein) (Figure 1-1) (10). ZnT and ZIP proteins are encoded by SLC30A and SLC39A family, respectively. ZnT family, comprising 10 members, facilitates zinc efflux from the cytosol to the extracellular space and lumens of intracellular compartments. ZIP family, consisting of 14 members, functions to mediate the zinc influx from the extracellular space or intracellular compartments into the cytosol. The substrate specificity of some ZIP transporters is not strict, which permits the transport of iron, manganese, and cadmium through the transporter protein. For instance, ZIP8 and ZIP14 have been characterized as transporters mediating the import of iron and manganese into cells, in addition to zinc (47–49). Zinc is required by fundamentally every live cell in humans, and the large number of zinc transporter genes allows cells to have cell-type-specific regulatory mechanisms for cellular zinc homeostasis and independently control their zinc pool in response to particular physiological cues or stress conditions. It also provides tissue-specific 7 mechanisms that contribute to the precise regulation of the distribution of the body zinc (50). For instance, ZIP4 is the zinc importer mediating the transfer of dietary zinc across the apical membrane and into the cytosol of absorptive enterocytes (51). The expression of ZIP4 is relatively specific to the small intestine, where the majority of dietary zinc is absorbed. This tissue-specific distribution of ZIP4 permits a targeted control of the zinc flux across the intestinal brush border when dietary intake levels of zinc fluctuate.When intake is restricted, ZIP4 abundance at the plasma membrane of the apical side of enterocytes increases which promotes the rate of zinc absorption. This compensatory upregulation of a zinc importer by zinc restriction may not occur in other cell types, or happen via response by a different zinc transporter gene, e.g., elevated ZIP10 for hepatocytes (52). A number of zinc transporter genes feature compensatory responses to cellular zinc status. In vertebrates, metal-response element-binding transcription factor-1 (MTF- 1) is the primary zinc-sensing factor and regulates the expression of multiple genes related to zinc homeostasis (53) (Figure 1-1). By elevated cytosolic zinc, MTF-1 and zinc form a complex, which translocate to the nucleus where it binds to metal response elements (MRE). MTF-1 binding to the MRE in the promoter region of the zinc exporter gene, ZnT1, and metallothionein (MT) results in transcriptional activation of the genes. MT is an intracellular metal-binding protein and carries 5-15% of the cytosolic zinc (10,54). Up to 7 atoms of zinc can be bound to MT with different affinities (54). MT functions in cytosolic zinc buffering and as a zinc donor for other zinc metalloproteins (54). On the other hand, MTF-1-MRE complexes downstream of the transcription start site can repress gene expression. When MTF-1 binds to the MRE of the zinc importer 8 gene, ZIP10, it interferes with polymerase II activity, which leads to transcriptional repression of the gene product (52). Systemic zinc homeostasis There is approximately 2-3 g of zinc in the adult human body, and serum zinc accounts for less than 0.1% of the total body zinc, which is predominantly bound to albumin (55). Intake of zinc is required on a daily basis because the body has no specialized zinc storage system (56). Although bone and skeletal muscle contain ~90% of the whole-body zinc, these are not considered a conventional zinc store since there is no way to augment the release of zinc during a deficiency. The systemic zinc homeostasis is achieved by regulating the absorption in the duodenum and jejunum and excretion via the gastrointestinal tract and kidney (57). Zinc absorption in the small intestine increases up to 90 % by a zinc-deficient diet (58). As mentioned earlier, ZIP4 expression in the small intestine rapidly responds to zinc depletion and repletion, implying its importance in controlling zinc absorption (59). ZIP4 protein is recruited to the apical membrane to facilitate the uptake of dietary zinc when scarce in the intestinal lumen. In a zinc-replete condition, the expression of Zip4 decreases, and the protein at the apical membrane is internalized and degraded via the ubiquitin-proteasome pathway (59). ZnT1, primarily located on the basolateral membrane of the enterocyte, facilitates the efflux of zinc (60). Dietary zinc supplementation also increases ZnT1 expression at both mRNA and protein levels, whereas zinc depletion does not affect the expression (60,61). Regulated excretion of endogenous zinc primarily occurs through the pancreas, which releases excess body zinc into the intestinal tract. This process involves three zinc transporter proteins, ZIP5, ZnT1, and ZnT2. ZIP5 is 9 predominantly distributed on the basolateral surface of the polarized pancreatic acinar cells (62). Similar to ZIP4 in enterocytes, ZIP5 initiates the removal of zinc by mediating the entry of zinc into the acinar cells and responds to body zinc status. ZnT1 and ZnT2, which exports acinar cell zinc through the apical membrane and into secretory vesicles, are downregulated by zinc deficiency to prevent unwanted losses of zinc (51,63,64). Unlike iron, zinc does not have a mechanism for regulated storage. Zinc in erythroid development and red cell metabolism Zinc and anemia Anemia is a major health concern worldwide, affecting approximately 23% of the population (65). The prevalence of anemia is high in women of reproductive age, children under five years old, and the Western Sub-Saharan African populations (66). According to the World Health Organization (WHO), anemia is defined as low hemoglobin levels or low red blood cell count. The association between zinc deficiency and anemia has been recognized in clinical and laboratory studies over the past several years. In the first study documenting zinc deficiency in humans, the symptoms of zinc deficiency included those similar to iron deficiency anemia (21,22). As a major predictor for hemoglobin concentrations and anemia risk, plasma and serum zinc have been suggested in previous clinical studies among pregnant women (2,3) and children (4,5), independent of iron status. In addition, zinc supplements improved hemoglobin response among young children with a risk of zinc deficiency when provided together with iron treatment for anemia (67). These indicate an association between zinc and the roles of iron in red cell homeostasis and hemoglobin production. 10 Several animal studies support the involvement of zinc in erythrocyte production and functioning. Apparent zinc assimilation in bone marrow, an erythropoietic tissue, has been reported in rats undergoing recovery from hemolytic anemia (6). Moreover, redistribution of zinc was observed in anemic mice and rats, resulting in decreased bone and plasma zinc and increased bone marrow and erythrocyte zinc (68). These indicate the importance of zinc supply to sites of red blood cell production when increased erythropoiesis occurs. Zinc has also been identified as important for the maintenance of red blood cells after maturation and in circulation. Dietary zinc restriction shortened the lifespan of erythrocytes due to membrane fragility in rats and pigs (35,36), impaired handling of oxidative stress in rats (69,70), and decreased erythropoietin synthesis in rats (71). Recently, a new molecular role of zinc as a molecular switch for red blood cell development has been identified (8). An accumulation of intracellular zinc early in terminal erythroid differentiation supports cell survival and is accomplished via GATA1- induced Zip8 expression. GATA1 is one of the master regulators for erythroid gene expression, and will be discussed again later in this Chapter. As heme accumulates upon the differentiation, the zinc importer Zip8 expression declines and the exporter ZnT1 becomes activated (8). This switch in the zinc transporter activities drives a drop in the intracellular zinc pool, which is thought to further support hemoglobinization and cellular differentiation (8). While these studies identify the regulatory role of zinc in erythroid differentiation, how cellular zinc status or deficiency influences red cell development or processes of erythroid maturation remains to be addressed. 11 Erythropoiesis Erythropoiesis is the formation of red blood cells. Committed erythropoiesis, also termed terminal erythroid differentiation, is the process after erythropoietin (EPO) stimulation, which induces the synthesis of erythroid-specific proteins and remodeling of the cellular structure to mature into erythrocytes (72) (Figure 1-2). In utero, the yolk sac is the primary organ for red blood cell formation, which is taken over by the liver later during fetal development. After birth, the bone marrow becomes the major site of erythropoiesis. Extramedullary erythropoiesis in the spleen and liver occurs under certain pathological conditions requiring a rapid expansion in red cell production, such as acute hemolytic anemia (6). Erythropoietic gene regulations EKLF has a multifunctional role in erythropoiesis as a zinc-finger transcription factor (12). During the hematopoiesis, EKLF promotes the differentiation of megakaryocyte-erythroid progenitor (MEP) towards erythroid lineage by inducing and coordinately regulating various erythropoietic genes (12). In contrast to most differentiation which occurs with cell-cycle arrest, terminal erythroid differentiation is accompanied by 3-4 cell cycle proliferation and reduced cell size and gene expression. EKLF plays a role in the regulation of cell cycle machinery (73). Additionally, the transcription factor activates genes for the enucleation of reticulocytes, which is the final process for maturation into red blood cells (12). GATA1 is another zinc-dependent transcription factor that orchestrates the expression of genes integral to committed erythroid development. GATA1 contains two zinc finger domains (74). GATA1 protein predominantly exists in the cytoplasm in the 12 progenitor cells, and then immediately translocates to the nucleus once the erythropoietin stimulates the progenitor to differentiate to proerythroblast (75). Nuclear GATA1 binds to a specific DNA motif, (A/T)GATA(A/G) (WGATAR), via its two C-terminal zinc- finger domains. A high abundance of GATA1 is necessary for the EPO-dependent early stages of erythroid development, whereas its expression declines at the late stages of differentiation (76,77). Zinc-containing metalloenzyme in mature red cell Another crucial role of erythrocytes, besides oxygen transport throughout the body, is in the transport of carbon dioxide for removal from the body via the respiratory system. Carbonic anhydrase mediates the conversion of carbon dioxide and water into carbonic acid (78). Among the 14 isozymes of carbonic anhydrases, carbonic anhydrase I (CA1) is the major form present in erythrocytes. The bicarbonate produced by erythrocytic CA1 activity is estimated to account for approximately 70% of the total carbon dioxide transport by the blood. In this process, a zinc ion functions as a catalytic component in the active site of CA1 (79). Differentiating erythroid progenitors experience a massive expansion in their cellular iron contents, which is required for heme biosynthesis (80). In addition, erythrocytes gradually release iron from their heme contents while in circulation (81). Iron atoms are redox-active and thus can impose oxidative stress on cells. Superoxide dismutase (SOD) plays a vital role in defending against oxidative stress and removing cytotoxic reactive oxygen species (ROS). SOD catalyzes the dismutation of the ROS, including superoxide (𝑂∙ି), to relatively stable molecules, such as hydrogen peroxide and 13 oxygen (13). There are three different forms of SOD in mammals: cytosolic Cu/Zn- dependent (SOD1), mitochondrial Mn-dependent (SOD2), and extracellular Cu/Zn- dependent SOD (SOD3). Erythrocyte is one of the cell types where SOD occurs in a high concentration, particularly zinc-containing SOD1. Heme biosynthesis The primary function of erythrocytes is to transport oxygen in the body via its hemoglobin contents (82). Hemoglobin contains four globin subunits (α1, α2, β1, and β2), each with a heme molecule bound. Heme is composed of the porphyrin ring, an organic ring-structure compound, with an iron atom. Hemoglobin forms a reversible bond with oxygen, which is called oxyhemoglobin, exhibiting red coloration. The mammalian heme biosynthesis pathway includes the following key reactions mediating protoporphyrin IX (PPIX) synthesis and the final incorporation of iron into PPIX for heme production (Figure 1-3). A mitochondrial enzyme, 5-aminolevulinic acid synthase2 (ALAS2), is the rate- limiting enzyme of heme biosynthesis (82). ALAS2 catalyzes the production of δ-ALA from glycine and succinyl-CoA in the mitochondria. ALA is transported to the cytosol, where it is used for the synthesis of PBG by ALAD. ALAD requires zinc as a cofactor, i.e., a catalytic component. Subsequently, the organic ring structure of heme, PPIX, is produced. The final step of heme biosynthesis involves the incorporation of an iron atom into the PPIX and is mediated by ferrochelatase (FECH) in the mitochondria. Notably, FECH can use zinc, instead of iron, as a substrate to generate zinc protoporphyrin (ZnPP) (83). Elevated ZnPP to heme ratio reflects an increase in cellular zinc to iron ratio in 14 developing red cells, and it has been used as an index of iron deficiency in clinical settings (84). Zinc transporter expression in the erythron While biochemical studies have revealed the roles of zinc in various processes of developing and mature red cells, the molecular mechanisms by which erythroid zinc is precisely regulated remain largely unknown. Among the 24 zinc transporters, the expressions of ZnT1, ZIP8, and ZIP10, have been detected in plasma membrane fractions of erythrocytes, i.e., ghost cells (85). The temporal regulation of these transporter genes, with the expression of importers preceding that of ZnT1, indicates their regulatory roles for cellular zinc homeostasis during terminal erythroid differentiation (85). Additionally, the erythrocytic ZnT1 and ZIP10 levels responded to dietary restriction in mice, suggesting their involvement in compensatory mechanisms for maintaining erythroid zinc homeostasis under zinc restriction (85). More recently, zinc import by ZIP8 at the early stages of terminal erythroid differentiation has been identified as essential for the cell survival of developing erythroid cells (8). Moreover, enhanced ZnT1 expression at the later stages of development was confirmed to determine the rate of erythroid differentiation by loss-of-function studies (8). The requirement of normal ZIP8 activity has been confirmed in vivo by a mutant mouse model of Zip8 knockdown, which features severe anemia and tissue gene expression profiles indicative of impaired hematopoiesis (86). Yet, the precise roles and requirements of ZIP10 in terminal erythroid differentiation remain uncharacterized. 15 Metal ions in nutritional immunity Hypoferremia and hypozincemia by inflammation Inflammation and infection lead to acute decreases in the circulating concentrations of trace elements, including iron, manganese, copper, and zinc (87,88). Inflammation-induced hypoferremia and hypozincemia are mediated by the hepatic regulation of transmembrane metal transporters, ferroportin (FPN), and ZIP14, respectively (89). Pathogenic microorganisms also require metal nutrients for survival, maturation, and proliferation. Thus, the decline in blood metal concentrations can limit further disease progression by producing a nutrient-poor environment of pathogens. This innate immune response, which aims to restrict the availability of nutrients to invasive pathogens, is termed nutritional immunity. Nutritional immunity by hypoferremia is produced by elevated hepcidin production and activity. Hepcidin, a hormone encoded by the HAMP (hepatic antimicrobial peptide) gene, is the central regulator of systemic iron homeostasis. The primary function of hepcidin is to repress the activity of FPN, the sole known iron export channel encoded by the SLC40A1 gene, on the plasma membrane of hepatocytes, splenic macrophages, and duodenal enterocytes. Hepcidin facilitates the internalization and lysosomal degradation of FPN in order to limit the flux of iron from these cells into circulation (90). In systemic inflammation status, hepcidin synthesis is induced by interleukin 6 (IL-6), which is a pro-inflammatory cytokine involved in the regulation of acute-phase responses (91). By elevated hepcidin, iron becomes sequestered in the target cell types, leading to hypoferremia (92). Conversely, hereditary hemochromatosis, an 16 inherited iron-overload disorder due to impaired hepcidin functioning, can potentially increase susceptibility to infections (93). Likewise, iron supplementation could lead to impairment of an innate immune system against infections in low- or normal-iron individuals (94). Functional zinc restriction by genetic mechanisms for redistribution of systemic zinc can provide protection against the growth and survival of invading pathogens (95). Lipopolysaccharide (LPS) produces responses of endotoxemia. In LPS-injected mice, the hepatic transcripts and protein expressions of metallothionein and Zip14 were greatly upregulated by IL-6 signaling, leading to hypozincemia (89,96). Moreover, Il6-knockout mice lack ZIP14 responses to endotoxemia (49). This upregulation of hepatic ZIP14 induced by inflammatory stimulus mediates the sequestration of zinc by the liver and restricts the supply of zinc to pathogens in the circulation. It is believed to serve as a primary mechanism for nutritional immunity against infection. Zinc availability at microenvironments of infection can also be restricted by proteins produced by immune cells. Neutrophils secrete calprotectin, an antimicrobial zinc-binding protein. For instance, calprotectin is released at abscesses area to chelate zinc and inhibit zinc- dependent microbial growth (97). Zinc transporter variants in diseases Acrodermatitis enteropathica (AE) is an inherited disease due to the mutation of the zinc transporter gene, ZIP4. Molecular and genetic studies on the etiology of AE revealed the physiological importance of ZIP4 in the absorption of dietary zinc, and thereby controlling systemic zinc homeostasis (59). AE patients present the symptoms of severe zinc deficiency, including dermatitis, impaired growth, and compromised immune 17 and reproductive systems. Notably, oral administration of high doses of zinc can effectively reverse symptoms of AE (98). These demonstrate how genetic variations can affect an individual’s dietary requirements of a particular nutrient for maintenance of a healthy state. The recent advances in genome sequencing technology have improved the quality, size, and accessibility of datasets that permit computational predictions of associations between genetic variations and disease risks. Through these resources, associations between zinc transporter variants and various diseases have been discovered and characterized. For instance, a missense variant of ZnT8 (rs13266634, R325W, risk C allele) in the pancreatic beta cells has been associated with type2 diabetes (99). Subsequent studies revealed the role of ZnT8 in the inhibition of hepatic insulin clearance by suppressing insulin endocytosis (18). Additionally, risks of other diseases have been attributed to mutations in zinc transporter genes, including spondylocheiro dysplastic Ehlers-Danlos syndrome by ZIP13 variants (100), autosomal recessive agammaglobulinemia by ZIP7 variants (101), transient neonatal zinc deficiency, and hepatic cirrhosis by mutations in ZnT2 and ZnT10, respectively (102,103). Metal transporter variants in immunity against endemic infections Malaria is a fatal infectious disease caused by Plasmodium parasites infected through mosquito bites. Globally, the population of the African continent accounts for approximately 92% of malaria incidence and deaths (104). As other pathogens, malarial parasites require iron and zinc for survival and growth. Recently, a mutation conferring protection against malaria in the gene encoding the iron exporter ferroportin (FPN) was discovered (105). The missense polymorphism in the FPN gene results in a Gln248His 18 substitution of the transporter protein, and is a common polymorphism in the African population compared to the other populations. Individuals with the FPN Gln248His mutation showed a protective impact against malarial infection, such as a longer duration of fever but less severe symptoms, compared to common allele subjects when they were exposed to uncomplicated malaria (81). The proposed mechanism of protection against malaria by the FPN mutation is the uncontrolled removal of nonheme iron within erythrocytes, resulting in less iron supply in erythrocytes to the pathogens growing within erythrocytes (81). Zinc is among the primary nutrients targeted by mechanisms of nutritional immunity (95). As noted earlier, ZIP4 is the zinc transporter mediating the absorption of dietary zinc through enterocytes. Recently, it has been suggested that ZIP4 variants unique in the Sub-Saharan African population may potentially confer nutritional immunity by producing mild zinc deficiency (106). An African-specific Leu372Val polymorphism of ZIP4 features less efficiency in zinc transport activity and thus is expected to cause less zinc absorption from the diet when compared to that by ZIP4 encoded by the common allele (106). Accelerated zinc accumulation occurs in erythrocytes invaded by the malaria parasite, implying an exceptional demand for zinc by the malaria parasites during their life stage within red blood cells (107,108). The role of host zinc transporters on erythrocyte membranes in this acute erythrocytic zinc accumulation has been postulated (108), but it has not been tested. The erythroid zinc importers ZIP8 and ZIP10 would be of particular relevance to this hypothetic model of zinc acquisition by infected red cells and the parasites therein (85). 19 Hypothesis and study aims The overall goal of this thesis was to determine the roles and regulation of zinc and its transporter gene, ZIP10, during red cell development. In addition, the study aimed to identify ZIP10 genetic variants specific to the African population, potentially affecting the functions of ZIP10. The overarching hypothesis of the studies was that the intracellular zinc pool of erythroid cells is regulated by differential ZIP10 activities. Each chapter tested a different aim. Aim 1: To determine the physiological roles of zinc during terminal erythroid differentiation. Several proteins integral to erythroid development and functioning require zinc as a structural or catalytic component, i.e., ALAD. The study focused on how zinc deficiency impairs proper heme biosynthesis using extensively validated in vitro models of terminal erythroid differentiation. Aim 2: To characterize the roles and regulation of erythroid ZIP10 during red cell development and zinc restriction. The expression of ZIP10 on the plasma membrane of mature red blood cells responds to dietary zinc levels in mice (85). Yet, the physiological implications of such ZIP10 response during red blood cell development remain unknown. Thus, the study aimed to assess the roles of erythroid ZIP10 during red cell development and zinc restriction by loss-of-function approaches employing siRNA-mediated gene silencing. The hypothesis was that loss of ZIP10 leads to functional zinc deficiency and exacerbates the consequences of limited zinc supply in erythroid cells. Aim 3: To identify ZIP10 genetic variants enriched in the African population. Screening through the 1000 Genome Project Phase 3 and the Genome Aggregation 20 Database (gnomAD) database of genetic variants revealed ZIP10 genetic variants uniquely present in the African population. This study aimed to experimentally validate the predicted prevalence of each ZIP10 genetic variant from African populations. 21 Figure 1-1. Location and direction of zinc transporters and a scheme of transcriptional regulation via MTF-1 in the nucleus. The location of zinc transporters and the directions of zinc mobilization are shown. Red arrows indicate ZIP activities, which are to facilitate zinc transport into the cytosol. Blue arrows indicate ZnT activities, which are to remove zinc from the cytosol. MTF-1-mediated transcriptional activation (bottom left) and suppression (bottom right) of genes involved in zinc homeostasis are shown. ** Abbreviations: ZnT, zinc transporters; ZIP, Zrt-,Irt like protein; ER, endoplasmic reticulum; MTF-1, metal-response element-binding transcription factor-1; MRE, metal response elements; Pol II, polymerase II; Mt, metallothionein. 22 Figure 1-2. A scheme of terminal erythroid differentiation. Terminal erythropoiesis, also termed terminal erythroid differentiation, is the process after erythropoietin (EPO) stimulation, which induces the synthesis of erythroid-specific proteins and remodeling of the cellular structure to mature into erythrocytes (red blood cells). 23 Figure 1-3. A scheme of heme biosynthesis in the cytosol and mitochondria. Heme is composed of the porphyrin ring, an organic ring-structure compound, with an iron atom. The mammalian heme biosynthesis pathway includes a mitochondrial enzyme, ALAS2, a cytosolic enzyme, ALAD, and FECH. ALAS2 catalyzes the production of ALA from glycine and succinyl-CoA in the mitochondria, then ALA is transported to the cytosol, where it is utilized for the synthesis of PBG by ALAD. The final reaction of heme biosynthesis is the incorporation of an iron atom into the PPIX by FECH in the mitochondria. **Abbreviations; ALAS2, aminolevulinic acid synthase-2; δ-ALA, aminolevulinic acid; ALAD, δ-aminolevulinic acid dehydratase; PBG, porphobilinogen; CO III, coproporphyrinogen III; PPIX, protoporphyrin IX; FECH, Ferrochelatase. 24 Chapter 2 . Zinc is required for heme synthesis in developing erythroid progenitors 25 Introduction Zinc is an indispensable trace element involved in diverse biological processes and serves as a structural, catalytic, and regulatory component for metalloproteins in erythroid metabolisms (9). During red cell development, zinc-containing proteins and enzymes play essential roles. GATA1 and EKLF are zinc-finger erythroid transcription factors, which function as master regulators of erythropoietic genes (109,110). Additionally, ALAD, the enzyme mediating the conversion of δ-ALA to PBG during heme biosynthesis, requires four zinc atoms as catalytic cofactors to be active in its octameric form (10,16,111,112). Zinc deficiency has been associated with anemia in animals and humans (21,22,35,36). The first human study documenting zinc deficiency reported that the symptoms of zinc deficiency included those similar to iron deficiency anemia (21,22). This observation has been supported by several animal studies indicating an involvement of zinc in erythrocyte production and functioning. Dietary zinc restriction shortened the lifespan of erythrocytes by increasing the fragility of the red cell membrane in rats and pigs (35,36) and impairing defense against oxidative stress in rats (69,70). Moreover, enhanced zinc acquisition in bone marrow was observed in rats undergoing recovery from hemolytic anemia (6), and redistribution of zinc was observed in anemic mice and rats, resulting in reduced bone and plasma zinc assimilation and increased flux of zinc into the bone marrow and the erythrocyte pool (68). These indicate that there is a physiological need for zinc at sites of erythropoiesis when the organismal erythropoietic demand rises. 26 Recently, a new role of intracellular zinc as a molecular switch for red blood cell development, particularly during terminal erythroid differentiation, has been identified (8). Once erythroid progenitors are committed to differentiate into erythrocytes, they undergo two phases of terminal erythroid differentiation (113). The first phase is EPO- driven and involves the production of key proteins mediating heme biosynthesis and other cellular changes of red cell maturation. During the second phase, cells carry out the production of large amounts of heme, which is iron-dependent (113). Employing a multi- omics approach, zinc transporters ZIP8 and ZnT1 were identified as regulatory targets of GATA1 and heme during terminal erythroid differentiation (8). Moreover, these zinc transporter responses were demonstrated as mechanisms mediating the early accumulation of intracellular zinc supporting cell survival, and the later removal of zinc which could potentially interfere with the final step of heme production (8). Ferrochelatase mediates the incorporation of iron into PPIX to form heme; however, zinc can become a substrate of this enzyme and lead to accumulation of ZnPP instead of heme, when in excess (114,115). While these findings demonstrate how the erythroid zinc pool might be under tight physiological control during terminal erythroid differentiation, how cellular zinc status determined by the extracellular zinc availability influences the efficiency of red cell development or heme production remains to be addressed. The aims of this chapter were to mechanistically characterize the roles and requirements of zinc in erythroid heme biosynthesis and to determine the physiological implications of zinc deficiency in anemia attributed to impaired erythroid hemoglobinization. G1E-ER4 cells, an erythroid progenitor cell line, were employed as 27 the primary study model for terminal erythroid differentiation, and MEL-DS19 cells, a mouse erythroleukemia cell line, were introduced as a complementary model of induced heme biosynthesis. The potential mechanisms by which zinc restriction affects the synthesis of heme and PPIX in developing erythroid progenitors were investigated through a combination of molecular, biochemical, and cell biology tools. Materials and Methods Maintenance of cells and induction for differentiation G1E-ER4 is a mouse erythroid cell-line, which undergoes erythroid differentiation at the presence of EPO and β-estradiol (β-Est) (116). G1E-ER4 cells were cultured in Iscove’s Modified Dulbecco’s Medium (IMDM) supplemented with 15% fetal bovine serum, 100 U/mL penicillin-streptomycin, monothioglycerol (1:10,000), 2 U /mL EPO, and conditioned medium from Kit ligand-producing Chinese hamster ovary (CHO) cells (1:200). Cells were maintained at 0.2-1.0 x 106 cells/mL and incubated in 5% CO2 at 37°C. To induce the differentiation, cells at a density of 0.2 x 106 cells/mL were treated with 100 nM β-Est for up to 48 hours. MEL-DS19 cell is a mouse erythroleukemia cell line with the clone DS19, which undergoes differentiation by dimethylsulfoxide (DMSO). MEL-DS19 cells were grown in the Dulbecco's Modified Eagle Medium (DMEM), including 10% heat-inactivated FBS, 2 mM glutamine, and 1% MEM non-essential amino acids solution (MEM NEAA) (117). Cells were maintained at 0.05-0.8 x 106 cells/mL and incubated in 5% CO2 at 37°C. The cells were seeded at a density of 5 x 104 cells/mL and cultured with 1.5% DMSO for up to 5 days. 28 Zinc treatments Zinc restriction was produced by adding diethylenetriamine pentaacetate (DTPA), an extracellular zinc-chelating agent, to the culture medium at the final concentrations of 10 to 50 μM up to 48 hours. The concentrations were determined based on the previous studies (52,118). The replenishment of zinc deficiency was achieved by supplementing equimolar zinc chloride to the culture medium. Major minerals and trace elements analyses Approximately 10 million cells were washed with ice-cold PBS containing 10 mM EDTA twice at 130 x g for 10 min. Cells were transferred to acid-washed tubes at the third washing step. Pelleted cells were digested with nitric acid overnight at room temperature (RT) and an additional 2 hours at 75 °C. The cellular mineral profiles were determined by inductively coupled plasma-mass spectrometry (ICP-MS). The contents were normalized to the total protein contents of each sample. Total cell counts and cell viability Trypan blue exclusive staining was conducted to assess total cell counts and viability using a hemocytometer or automated cell counter R1 (Olympus). Cell suspensions were mixed with an equal volume of 0.4% trypan blue to stain dead cells prior to counting. RNA isolation, Reverse Transcription, and quantitative RT-PCR Reverse transcription and real-time quantitative polymerase chain reaction (qPCR) were conducted to measure the relative expression of the transcripts. Cells were harvested by centrifugation at 600 x g, 4°C for 5 min with ice-cold phosphate-buffered saline (PBS) washing. To isolate total RNA, cells were lysed in TRI reagent (Sigma- 29 Aldrich). RNA was isolated using the Direct-zol RNA mini preparation kit (Zymo Research) or 1‑bromo-3‑chloropropane according to the TRI Reagent manufacturer’s protocol. The equal amounts of extracted total RNA (250 ng/reaction) were transcribed into cDNA using the High-Capacity cDNA Reverse Transcription kit (Applied Biosystems) according to the manufacturer’s protocol. Transcripts were PCR-amplified using Power SYBR Green Master Mix (Applied Biosystems) and detected by CFX Connect Real-Time System (Bio-Rad). Relative mRNA abundance was determined by the 2−ΔΔCt method, with Tbp as a housekeeping gene for normalization. Primers for qPCR amplification are listed in Table 2-1. Western analyses Cells were harvested by centrifugation at 600 x g, 4°C for 5 min with ice-cold phosphate-buffered saline (PBS) washing and lysed in the Pierce RIPA buffer (Thermo Fisher Scientific) supplemented with a protease inhibitor cocktail. The lysate was centrifuged at 12,000 x g, 4°C, 5 min to remove cell debris, and the supernatant was collected for subsequent assays. Total protein concentrations were determined using the Pierce BCA Protein Assay kit (Thermo Fisher Scientific). For immunoblotting, an equal amount of protein (15-45 μg) was prepared with 2.5% β-mercaptoethanol (BME) and separated by Bis-Tris polyacrylamide gel electrophoresis (PAGE) in sodium dodecyl sulfate (SDS) running buffer. The separated protein was transferred to the nitrocellulose membrane using the Trans-Blot Turbo Transfer System (Bio-Rad). The efficient transfer was confirmed by Ponceau staining. Immunoblotting was conducted by primary incubations with mouse anti-MT1 (1:1,000; Invitrogen, MA 1-25479), rabbit anti-ALAD (1:1,000; Invitrogen, PA5-88521), rabbit anti-hemoglobin-alpha (HBA; 1:1,000; 30 Proteintech, 14537-1-AP), rabbit anti-IRP2 (1:1000; Dr. Betty Leibold, University of Utah), rabbit anti-ferritin (1:1000; Sigma-Aldrich, F5012), and anti-GAPDH (1:2,000; Bio-Rad, 12004167) primary antibodies. Respective protein expression was visualized using NIR Fluorescent 680 or 800 nm secondary antibodies (1:10,000, Li-Cor) and Li- Cor Odyssey Fc detection system. RNA-seq transcriptome analyses and data processing Total RNA from approximately five million G1E-ER4 cells were extracted using Direct-zol RNA MiniPrep with DNase I treatment. RNA yield and purity were determined using Nanodrop-2000 spectrophotometer (Thermo Fisher). RNA integrity was assessed using a Bioanalyzer, and the RNAs with RNA integrity number (RIN) > 8.0 were processed for libraries preparations and sequencing by the University of Minnesota Genomics Center (UMGC). The RNA sequencing was performed with 20 million reads per sample for 150 bp paired-end run by NovaSeq 6000 (Illumina). Post-hoc bioinformatic analyses of normalized sequencing data were conducted using volcano plot (119) and the ingenuity pathway analysis (IPA) tools (120). To define the similarity between the groups, hierarchical cluster analysis of differentially expressed genes (DEG) by zinc deficiency was performed using average linkage through Morpheus (https://software.broadinstitute.org/morpheus/). Hierarchical clustering of gene expression profile was based on log10(TPM+1) and normalized to z-score using three biological replicates for −DTPA and +DTPA at 24-hour groups and one sample for the 0- hour group. 31 Heme, ZnPP, and total PPIX assays Cellular heme contents were colorimetrically measured (121). Washed cells were lysed in the QuantiChrome heme assay solution at RT for 5 min, and absorbance at 400 nm was measured. Heme contents were normalized to total cell counts or protein concentrations. To measure total PPIX contents, the equal number of cells were lysed in a lysis buffer containing a protease inhibitor cocktail and centrifuged at 12,000 x g, 4℃, 10 minutes. The supernatant was added to ethyl acetate and acetic acid (4:1) and centrifuged at 1,000 x g for 3 min to remove precipitated proteins. The supernatant was extracted in 1.5N HCl by centrifugation at 1,000 x g for 3 min. The extracted PPIX was transferred to a black 96-well plate. The fluorescence was measured using a spectrofluorometer at an excitation wavelength of 405 nm and an emission wavelength of 604 nm (122–124). To determine the ZnPP and free PPIX, the equal number of cells in Tri-HCl buffer were lysed using homogenizer through centrifugation at max speed, 2 min at RT. The lysates were mixed with pure ethanol (1:9 ratio). The mixture was vortexed and centrifuged for 10 min at 20,000 x g. The supernatants were transferred to a black 96- well plate, and the fluorescence was measured using a spectrofluorometer at an excitation wavelength of 415 nm and an emission wavelength of 589 nm and 623nm for ZnPP and free PPIX, respectively. Statistics All in vitro experiments include at least three biological replicates or independent experiments. Data were presented as mean ± standard deviation. Paired t-test, Student’s t- test, one-way ANOVA, two-way ANOVA, or repeated-measures ANOVA followed by 32 Dunnett’s or Tukey's HSD post-hoc test was conducted to define significant differences. A P-value < 0.05 was considered statistically significant. 33 Table 2-1. Primers for SYBR Green-based qPCR analyses Gene of Interest Primer direction Primer sequence Alas2 Forward 5’ - CAGAGGGCAGCTCCAGAAGTT - 3’ Reverse 5’ - GCTTCGGGTGGTTGAATCC - 3’ Alad Forward 5' - GCTCAGTCAAGCCCAGCTTT - 3' Reverse 5' - CTTGAATGTCTCGGGCCACTG - 3' Hba-a1/2 Forward 5' - CGTGCTGACCTCCAAGTACC - 3' Reverse 5' - GGTACAGGTGCAAGGGAGAG - 3' Hbb-b1 Forward 5’ - ACTGCCCTGGCTCACAAGTA - 3’ Reverse 5’ - ACCATTGTTCACAGGCAAGAGC - 3’ Mt1 Forward 5' - CCTCCTGCAAGAAGAGCTGC - 3' Reverse 5' - TTCGTCACATCAGGCACAGC - 3' Tfrc Forward 5’ - TCACTTCCTGTCGCCCTATGT - 3’ Reverse 5’ - AGAGTGTGAGAGCCAGAGCC - 3’ Gata1 Forward 5' - CAAGCTCCATCAGGTGAACCG - 3' Reverse 5' - TTCCCTTTGCCAGATGCCTTG - 3' Abcb6 Forward 5' - CGTCATAGCACACAGGCTCTC - 3' Reverse 5' - TGCAACCGTCCTTGATGACC - 3' Mfrn1 Forward 5' - AATGAATCCAGCAGAAGTGGTGA - 3' Reverse 5' - ACAACTGAAGGCTGACTGGTG - 3' Tbp Forward 5' - AGTTGTGCAGAAGTTGGGCT - 3' Reverse 5' - TACTGAACTGCTGGTGGGTCA - 3' Actb Forward 5' - AGGAGTACGATGAGTCCGGC - 3' Reverse 5' - AGCTCAGTAACAGTCCGCCT - 3' 34 Results Differentiation of G1E-ER4 cells First, the differentiation of G1E-ER4 cells was confirmed by measures of erythropoietic gene responses. The temporal patterns of relevant transcript abundances were measured after β-estradiol treatment (121). Positive controls of the terminal erythroid differentiation included Alas2 and Alad, which are enzymes involved in PPIX synthesis, and alpha- and beta-globin genes. The mRNA abundance of Alas2, Alad, Hba- a1/2, and Hbb-b1 markedly increased 24 hours after induction of differentiation (Figure 2-1). These confirmed that our G1E-ER4 cell culture model appropriately underwent the terminal erythroid differentiation upon induction by β-estradiol. The cellular zinc pool expands during the terminal erythroid differentiation Although a number of studies have reported that iron is exceptionally accumulated in developing erythroid progenitors because of its essentiality for producing heme (121,125), there remain many unknowns regarding other minerals in red cell development. Thus, to determine whether the red cells require other exogenous minerals during differentiation, the cellular contents of major minerals including sodium, magnesium, phosphorous, potassium, and calcium, and trace elements including manganese, iron, copper, and zinc were analyzed via ICP-MS. Phosphorous was measured as an internal negative control. After 48 hours of differentiation, the significant accumulations of sodium, potassium, calcium, manganese, iron, copper, and zinc contents were observed in developing G1E-ER4 cells, suggesting the demands for these exogenous minerals in terminal erythroid differentiation (Table 2-2). Interestingly, G1E- 35 ER4 cells featured a 1.7-fold increase in total cellular zinc contents at 48 hours of differentiation, and the post-development cellular contents of zinc were measured at a level comparable to those of iron. This finding identifies the substantial demand for zinc during red blood cell development. Zinc is required for hemoglobinization Since the accumulation of cellular zinc in differentiating G1E-ER4 cells was observed via ICP-MS-based metal quantitation, we next tested whether restriction of cellular zinc supply impairs erythroid development. Three different doses of DTPA (10, 25, and 50 μM) were treated for 48 hours to limit the availability of extracellular zinc to developing G1E-ER4 cells. While the lower doses did not impair the red coloration of developing G1E-ER4 cells, zinc restriction by 50 μM DTPA led to less intense red coloration, implying impaired hemoglobinization (Figure 2-2 A). Less heme production by the higher dose of DTPA was confirmed by quantitative analyses of cellular heme contents, as shown in Figure 2-2 B. Additionally, indices of impaired heme production were observed by a shorter 24-hour DTPA treatment at 50 μM (Figure 2-3). Zinc deficiency has been associated with cell death, such as apoptosis, in leukemia cells and macrophages (126,127), and cell death could impact hemoglobin production. Thus, we tested to determine whether zinc restriction by DTPA affects the total number of cells and cell viability. By 48 hours of the DTPA treatment, the total number and viability were decreased (Figure 2-4 A, B). However, by 24 hours of zinc restriction, when heme impairment was also observed, DTPA treatment did not reduce the total number of cells and cell viability, implying that impairment of hemoglobinization occurs prior to cell death and thus was not due to cell death. 36 Additionally, the cell size of DTPA-treated cells at 48 hours of the development was decreased, consistent with microcytosis of erythroid cells experiencing impaired heme production (Figure 2-4 C). DTPA may chelate other trace elements in the cell culture medium (118). To confirm that the impairment of hemoglobinization by DTPA is attributable to zinc restriction, DTPA and equimolar zinc chloride (50 μM) were treated to the differentiating G1E-ER4 cells. The color of cell pellets and normalized cellular heme contents were assessed 48 hours after differentiation and treatments. All indices of impaired hemoglobinization by DTPA treatment were fully restored by adding back equimolar zinc, identifying zinc restriction as the primary cause of DTPA-induced impaired heme production (Figure 2-5). In addition, cellular zinc changes by DTPA and zinc chloride were measured via ICP-MS-based metal analysis at 24 and 48 hours of the development. As expected, the cellular zinc contents reduced by DTPA returned to normal by the addition of equimolar zinc at both 24 and 48 hours (Figure 2-6). Previous studies suggest how zinc requirements may temporally change during erythroid development (8,85). Thus, whether the timing of zinc restriction can change its influence on erythroid heme synthesis was assessed. Notably, zinc restriction during the first 24 hours of differentiation led to impaired heme biosynthesis, while a delayed 24- hour treatment of DTPA did not affect hemoglobin production by differentiating G1E- ER4 cells (Figure 2-7). This implies an early role of zinc in terminal erythroid differentiation, which is essential for later heme biosynthesis. Hemoglobin consists of heme and globin, and heme interacts with the globin expression. Heme molecules positively regulate globin expression at the transcriptional 37 level via inhibition of BTB Domain And CNC Homolog 1 (BACH1) binding activity on the enhancer regions of globin genes (128). The translation of globin proteins can also be promoted by heme via inactivation of heme-regulated inhibitor (HRI) kinase (129). To test the effect of reduced heme contents by zinc restriction on globin expressions, mRNA and protein levels of alpha-globin were assessed in developing G1E-ER4 cells. Metallothionein mRNA and protein abundance was measured as biomarkers of zinc status. The restriction of cellular zinc supply was confirmed by lower MT expression and led to diminished mRNA and protein abundance of alpha-globin in the DTPA-treated cells during the development (Figure 2-8). Zinc deficiency does not change cellular iron contents but may affect the cellular distribution of iron. Heme production requires adequate levels of two key metabolites, iron and PPIX. Thus, impaired heme biosynthesis by zinc restriction could be attributable to either inadequate iron supply or PPIX synthesis. To test this, we first determined if DTPA resulted in changes in the cellular iron contents using ICP-MS. Total cellular iron levels did not change by DTPA treatment at either 24- or 48-hour of differentiation (Figure 2-9). Additionally, iron regulatory protein 2, IRP2, protein abundance and transferrin receptor, Tfrc, mRNA level were measured as indicators of the cytosolic labile iron pool. IRP2 is a posttranscriptional regulator of iron metabolism, which is degraded by expansion in the labile iron pool, and transferrin receptor mediates cellular iron uptake by endocytosis of circulating transferrin-bound iron, which is upregulated by reduction in the labile iron pool (130,131). These have been used as biomarkers of change in the cytosolic labile iron pool of erythroid progenitors (121). IRP2 protein abundance and Tfrc 38 mRNA level were not affected by DTPA at 24 hours of differentiation, demonstrating that impaired hemoglobinization by zinc restriction at this time-point was not due to a change in the cytosolic iron availability (Figure 2-10 A). However, at 48 hours of the differentiation, both IRP2 protein abundance and Tfrc mRNA level declined by DTPA, indicating an expansion in the cytosolic labile iron pool by zinc restriction (Figure 2-10 B). This could be due to an accumulation of cellular nonheme iron within the cytosol due to its impaired conversion to heme iron when the cellular zinc supply is inadequate. Transcriptome profile of zinc-deficient erythroid progenitors To obtain an unbiased and comprehensive understanding of zinc-restricted erythroid development, the transcriptome profile of differentiating G1E-ER4 cells with and without DTPA was compared using RNA-seq. A total of 1721 differential expressions by DTPA treatment were identified when the thresholds for significance were set at |Fold change| > 2.0 and false discovery rate (FDR) p-value < 0.05. A total of 3326 genes was determined when the criteria of |Fold change| > 1.5 and FDR p-value < 0.05 were applied. Hierarchical cluster analysis of differential expressions by zinc restriction using the average linkage clustering method discriminated cells differentiating under zinc restriction from those either proliferating or differentiating in control zinc- adequate medium (Figure 2-11). Among the DEGs by the criteria of |Fold change| > 2.0 and adjusted FDR P-value < 0.01, a heme biosynthetic gene, Alad, was significantly downregulated by DTPA, whereas Alas showed no significant response to zinc restriction (Figure 2-12). 39 Differential responses of heme biosynthetic genes to zinc restriction The RNA-seq data suggests the potential impairment in PPIX production due to lower expression of ALAD in zinc-restricted erythroid progenitors. To test this, the expression of ALAD and other heme biosynthetic genes were determined using qPCR and western blotting. The key enzymes and transporters mediating heme biosynthesis in the mitochondria and cytosol are presented in Figure 2-13 A. The first step of heme production occurs in the mitochondria by utilizing glycine and succinyl-CoA as substrates. ALAS2, located in mitochondria, is the first enzyme of heme biosynthesis, which produces δ-ALA. The δ-ALA is transported to the cytosol to be catalyzed by ALAD. ALAD is the first cytosolic enzyme in the heme biosynthetic pathway, which requires zinc atoms as catalytic cofactors (83). Among all the genes encoding enzymes for heme biosynthesis, Alad was the most responsive to zinc restriction (Figure 2-13 B). The RNA-seq data identified a decrease in Mt1 mRNA expression, confirming cellular zinc restriction by DTPA. While transcript abundance of Alas2, the gene for the first heme-biosynthetic enzyme in the mitochondria, was not affected by DTPA, mRNA level of Alad, which encodes the initial cytosolic enzyme for PPIX production decreased by DTPA treatment (Figure 2-14 A). These findings from the RNA-seq were further confirmed by qPCR results, which revealed no change in Alas2 and a decrease in Alad mRNA abundance. Moreover, ALAD protein abundance was decreased by DTPA at both 24 and 48 hours of differentiation (Figure 2-14 B). These collectively suggest that impaired heme biosynthesis by zinc restriction is not due to impaired induction of terminal erythroid differentiation (demonstrated by successful induction of Alas2) nor 40 inadequate iron import but rather caused by inadequate synthesis of PPIX mediated by ALAD. Protoporphyrin metabolism during zinc restriction. The findings from RNA-seq and quantitative expression analyses above suggest that zinc deficiency may impair the production of PPIX required for heme biosynthesis. Thus, the effects of zinc restriction on total PPIX contents in developing G1E-ER4 cells were investigated. In agreement with the lower ALAD production by zinc-depleted cells, the total cellular contents of total PPIX were significantly lower in cells differentiation with DTPA treatment (Figure 2-15). Cellular PPIX can be present in the form of heme, ZnPP, and free PPIX in the erythron. Thus, how cellular zinc status affects the abundance of ZnPP and free PPIX contents was assessed. For this part of the study, MEL-DS19 cells were employed because G1E-ER cells did not produce measurable levels of ZnPP or free PPIX (Figure 2-16). G1E-ER4 cells can differentiate from proerythroblasts into orthochromatic erythroblasts in vitro and thus only recapitulate the early stages of terminal erythroid development (121,132), which may explain the difficulty of detecting quantitatively meaningful levels of cellular ZnPP and free PPIX. MEL-DS19 cells can be induced to produce hemoglobin in the presence of DMSO (117). MEL-DS19 cell pellets became red, indicative of heme production, and accumulated heme and total PPIX after treatment with DMSO for up to 5 days (Figure 2-17 A, B, C). Moreover, increases in mRNA abundance of hemoglobin-synthetic genes, including Alas2, Hba-a1/2, and Hbb-b1, were measured in MEL-DS19 treated with DMSO for 3 days (Figure 2-18). This confirms that MEL-DS19 cells undergo the 41 establishment of the heme biosynthetic machinery and produce heme in response to DMSO as G1E-ER4 cells do by differentiation. MEL-DS19 cells continued accumulation of heme and total PPIX when cultured with DMSO for more than 3 days (Figure 2-17 B, C). Furthermore, the spectral peaks corresponding to ZnPP and free PPIX became prominent after 4 and 5 days of DMSO treatment, allowing quantitative assessments of these heme-related metabolites in the cells, whereas 3 days of DMSO treatment did not present peaks corresponding to ZnPP and free PPIX (Figure 2-16). As for heme and total PPIX, the spectra of MEL-DS19 cells at day 5 of DMSO treatment presented substantially higher ZnPP and free PPIX compared to those of MEL-DS19 at day 3 of DMSO treatment (Figure 2-17 D, E). To determine if zinc restriction results in less production of free PPIX and ZnPP, MEL-DS19 were treated with DMSO for 5 days along different lengths of DTPA. By the early induction of zinc deficiency (introduced by DTPA treatments on day 0, 1, 2, or 3 of DMSO treatment), both ZnPP and free PPIX contents in MEL-DS19 were significantly reduced (Figure 2-19). A delayed introduction of zinc restriction (on day 4 of DMSO treatment) did not produce changes in free PPIX and heme contents but did acutely repress the levels of ZnPP in the cells. Erythrocyte zinc protoporphyrin:heme (ZPP/H) ratio is a sensitive screening parameter of iron deficiency and iron deficiency anemia (84,133). In contrast to the rise in ZPP/H ratio by iron deficiency (84,133), zinc restriction acutely lowered the ZPP/H ratio by 46% compared to the control cells differentiating with adequate zinc. 42 Zinc deficiency does not globally impair GATA1-induced gene expression. GATA1 is a transcriptional regulator integral to the initiation of erythroid development and contains zinc-finger motifs requiring zinc as a structural component (109,110). G1E-ER4 cells are a genetically modified mouse erythroid cell-line, which undergoes development by restoring GATA1 in the presence of EPO and β-estradiol (116). The RNA-seq analysis of G1E-ER4 cells revealed a downregulation in Gata1 expression by DTPA treatment (Fold change: -2.274, FDR p-value < 0.001). To determine if this response of Gata1 to zinc restriction could produce a global impairment in GATA1-dependent gene expression, we analyzed the transcriptome profiles from RNA-seq using the upstream regulator analysis of the IPA (Qiagen). Despite a -2.274 fold change in Gata1 transcript abundance, the transcriptome profile of zinc-restricted G1E-ER4 cells revealed the mode of change in GATA1 activity by DTPA treatment to be positive with an activation z-score of 2.318 (p-value = 6.16 x 10-21). The activation z- score is indicative of the predicted activation state of the upstream regulator. The fold changes of GATA1-regulated genes identified by IPA in response to DTPA treatment are shown in Table 2-3. 43 Figure 2-1. Temporal patterns of transcripts during terminal erythroid differentiation. Transcripts of (A) Alas2, (B) Mt1, (C) Hba-a1/2, and (D) Hbb-b1 are presented upon β-estradiol treatment time. Genes of interest were normalized to Tbp. Independent experiments, n=3. Values represent mean ± SD. P-values were calculated by two-way ANOVA followed by Tukey’s HSD post-hoc test. *P < 0.05 versus −β-Est. **Abbreviation: β-Est, β-estradiol; Alas2, aminolevulinic acid synthase-2; Alad, δ-aminolevulinic acid dehydratase; Hba-a1/2, hemoglobin alpha adult chain 1 and 2; Hbb-b1, hemoglobin beta adult chain 1. 44 Table 2-2. The total cellular contents of minerals in proliferating and differentiating erythroid progenitors. The cellular contents of sodium, magnesium, phosphorous, potassium, calcium, manganese, iron, copper, and zinc in G1E-ER4 cells were analyzed using ICP-MS-based metal quantitation with or without adding β-estradiol. The contents were normalized to gram protein. Phosphorus was quantified as an internal control. Biological replicates, n=4. Values represent mean ± SD. *P < 0.05 by Student’s t-test. **Abbreviation: β-Est. β- estradiol. β-Est (48 h) Major minerals Trace elements Na / 23 Mg / 24 P / 31 K / 39 Ca / 40 Mn / 55 Fe / 56 Cu / 63 Zn / 66 (mmol /g Protein) (µmol /g Protein) (mmol /g Protein) (mmol /g Protein) (µmol /g Protein) (µmol /g Protein) (µmol /g Protein) (µmol /g Protein) (µmol /g Protein) − 1.66 ± 0.21 98.3 ± 5.8 1.43 ± 0.08 0.558 ± 0.036 1.56 ± 0.21 0.013 ± 0.001 0.76 ± 0.05 0.130 ± 0.010 3.45 ± 0.21 + 2.86 ± 0.22* 60.8 ± 2.3* 1.53 ± 0.07 0.614 ± 0.020* 6.34 ± 0.32* 0.105 ± 0.008* 5.10 ± 0.30* 0.455 ± 0.029* 5.87 ± 0.19* P-value 0.0002 <0.0001 0.1320 0.0323 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 45 Figure 2-2. Cellular zinc restriction impairs hemoglobinization in developing G1E-ER4 cells. (A) Representative pellet image of G1E-ER4 cells treated with different doses of DTPA (0, 10, 25, and 50 μM) at 48 hours of the differentiation is shown. The pellets were normalized to an equal number of cells. (B) Heme contents per cell of DTPA-treated cells were measured. Biological replicates, n=4. Values represent mean ± SD. P-values were calculated by one-way ANOVA followed by Dunnett’s post-hoc test. *P < 0.05 versus −DTPA. ** Abbreviations: β-Est, β-estradiol; DTPA, diethylenetriamine pentaacetate. 46 Figure 2-3. Zinc restriction impairs heme production from the early stage of terminal erythroid differentiation in G1E-ER4 cells. (A) Representative pellet images of zinc-deficient G1E-ER4 cells normalized to an equal number of cells are shown at both 24 and 48 hours of the differentiation. (B) Heme contents were measured at 24 and 48 hours of the differentiation. Biological replicates, n=3. Values represent mean ± SD. P-values were calculated by two-way ANOVA followed by Tukey’s HSD post-hoc test. *P < 0.05. **Abbreviations: β-Est, β-estradiol; DTPA, diethylenetriamine pentaacetate. 47 Figure 2-4. The total cell counts, cell viability, and live cell size are not affected by zinc restriction at 24 hours of G1E-ER4 cell differentiation but are reduced at 48 hours of differentiation. (A) Total cell counts, (B) cell viability, and (C) cell size were measured at 24 and 48 hours of the differentiation in G1E-ER4 cells using automatic cell counter R1 (Olympus). Biological replicates, n=4. Values represent mean ± SD. P-values were calculated by two-way ANOVA followed by Tukey’s HSD post-hoc test. *P < 0.05. **Abbreviations: β-Est, β-estradiol; DTPA, diethylenetriamine pentaacetate; n.s., not significant. 48 Figure 2-5. Decreased heme contents by DTPA treatment are fully restored by adding back equimolar zinc. (A) Representative pellet images of zinc-deficient cells and zinc-replenished cells normalized to an equal number of cells are shown. (B) Heme contents by zinc deficiency (50 µM DTPA) and zinc replenishment (50 µM DTPA and 50 µM zinc chloride) at 48-hour differentiation were measured. Biological replicates, n=4. Values represent mean ± SD. P-values were calculated by one-way ANOVA followed by Dunnett’s post-hoc test. *P < 0.05 versus CON. **Abbreviations: β-Est, β-estradiol; CON, control; DTPA, diethylenetriamine pentaacetate; Zn, zinc chloride; n.s., not significant. 49 Figure 2-6. Reduced cellular zinc contents by DTPA are fully replenished by adding back equimolar zinc. Cellular zinc contents at 24 and 48 hours of differentiation were measured using ICP-MS. The results were normalized to gram protein. Biological replicates, n=4. Values represent mean ± SD. P-values were calculated by two-way ANOVA followed by Dunnett’s post-hoc test. *P < 0.05 versus CON. **Abbreviations: CON, control; DTPA, diethylenetriamine pentaacetate; Zn, zinc chloride; n.s., not significant. 50 Figure 2-7. Adequate zinc supply is essential for heme biosynthesis at the early stage of the terminal erythroid differentiation. (A) A scheme of zinc deficiency by 50 μM of DTPA treatment is shown. The black period indicates that 50 μ of DTPA was treated, and the white period indicates either no DTPA treatment or 50 μ of DTPA with 50 μ of zinc chloride replenishment. (B) Representative pellet images of different treatment time points of DTPA in differentiating G1E-ER4 cells are shown. (C) Heme contents at 48-hour of the differentiation were measured. Biological replicates, n=3. Values represent mean ± SD. P-values were calculated by one-way ANOVA followed by Dunnet’s post- hoc test. *P < 0.05 versus the ̶ DPTA. * P < 0.05 by. **Abbreviations: DTPA, diethylenetriamine pentaacetate; ZnA, zinc-adequate; ZnD, zinc-deficient; n.s., not significant. 51 Figure 2-8. Restriction of zinc supply decreases alpha-globin mRNA and protein abundance in developing G1E-ER4 cells. (A) Transcript abundance of Mt1 and Hba-a1/2 was measured at both 24 and 48 hours of the differentiation. Genes of interest were normalized to Tbp. Biological replicates, n=3. Values represent mean ± SD. *P < 0.05 by two-way ANOVA followed by Tukey’s HSD post-hoc test. (B) Protein expressions of α-Globin and MT were measured. GAPDH was detected for normalization. Independent experiments, n=4. Values represent mean ± SD. P-values were calculated by two-way repeated-measures ANOVA followed by Tukey’s HSD post-hoc test. *P < 0.05 versus −DTPA. **Abbreviations: β-Est, β-estradiol; DTPA, diethylenetriamine pentaacetate; n.s., not significant. 52 Figure 2-9. Zinc restriction has no effect on total cellular iron contents during red cell development. Cellular iron contents at 24 and 48 hours of differentiation were measured using ICP-MS. The results were normalized to gram protein. Biological replicates, n=4. Values represent mean ± SD. P-values were calculated by two-way ANOVA followed by Dunnett’s post-hoc test. *P < 0.05 versus CON. **Abbreviations: CON, control; DTPA, diethylenetriamine pentaacetate; Zn, zinc chloride; n.s., not significant. 53 Figure 2-10. Cytosolic labile iron pool is not altered by DTPA treatment at 24-hour of differentiation but may be affected by DTPA at 48-hour of differentiation in G1E-ER4 cells. Western blot image and protein abundance of IRP2 and transcript abundance of transferrin receptor were measured at (A) 24 hours and (B) 48 hours of the differentiation. IRP2 protein was normalized to GAPDH. Independent experiments, n=3. Tfrc mRNA was normalized to Tbp. Biological replicates, n=3. P-values were calculated by paired t-test for IRP2 and by Student’s t- test for Tfrc. *P < 0.05. Values represent mean ± SD. **Abbreviations: β-Est, β-estradiol; DTPA, diethylenetriamine pentaacetate; n.s., not significant. 54 Figure 2-11. Hierarchical cluster analysis of differential expressions by zinc restriction discriminates zinc-deficient developing G1E-ER4 cells from zinc-adequate proliferating and developing G1E-ER4 cells. A comprehensive comparison between 0-hour of development G1E-ER4 (G0-C), zinc-adequate G1E-ER4 at 24 hours of differentiation (G24-C), and zinc-deficient (G24-D) G1E-ER4 cells at 24 hours of differentiation are presented. The analysis of gene expression based on log10(TPM+1), and normalized to z-score (Metric: One minus Pearson correlation; Linkage method: Average). Each cell represents the z-score normalized differentially expressed genes (DEGs) by DTPA treatment with (A) the criteria of |Fold change| > 1.5 and False Discovery Rate P-value < 0.05 and (B) the criteria of |Fold change| > 2.0 and False Discovery Rate P-value < 0.05. **Abbreviations; TPM, transcript per million. 55 Figure 2-12. Volcano plot of DEGs by zinc restriction in developing G1E-ER4 cells. Volcano plot of differential expressed genes (DEGs) of DTPA treatment comparison at 24 hours of differentiation in G1E-ER4 cells. Blue dots represent the down-regulated DEGs, and red dots represent the up-regulated DEGs by the criteria of |Fold change| ≥ 2.0 and adjusted false discovery rate (FDR) P-value < 0.01. Gray dots represent the genes of either |Fold change| < 2.0 or adjusted FDR P-value ≥ 0.01. 56 Figure 2-13. ALAD is the most responsive to zinc restriction among heme biosynthetic genes. (A) A scheme of heme biosynthesis and enzymes involved in the pathway. (B) A heat map shows the fold changes of TPM by 50 μM DTPA treatment at 24 hours of differentiation from RNA-seq data in log2 scale. * indicates significant DEGs determined by the criteria of |Fold change| ≥ 2.0 and adjusted false discovery rate (FDR) P-value < 0.01. **Abbreviations; ZnD, zinc deficiency; FC, fold change; TPM, transcript per million; ALAS2, aminolevulinic acid synthase-2; δ-ALA, aminolevulinic acid; ALAD, δ-aminolevulinic acid dehydratase; PBG, porphobilinogen; HMBS, hydroxymethylbilane synthase; HMB, hydroxymethylbilane; UROS, uroporphyrinogen I synthase; URO, uroporphyrinogen; UROD, uroporphyrinogen decarboxylase; CO III, coproporphyrinogen III; CPOX, coproporphyrinogen oxidase; TSPO, mitochondrial translocator protein; PPG, protoporphyrinogen; PPOX, Protoporphyrinogen oxidase; FECH, ferrochelatase; PPIX, protoporphyrin IX; SLC, solute-carrier gene; Fe, iron. 57 Figure 2-14. Zinc restriction does not impair Alas2 expression but represses Alad mRNA and protein expression. (A) Transcript abundance of Alas2 and Alad was measured at 24 hours of the G1E-ER4 differentiation. Genes of interest were normalized to Tbp. Biological replicates, n=3. Values represent mean ± SD. P-values were calculated by Student’s t-test. *P < 0.05. (B) Protein expressions of ALAD. ALAD protein was normalized to GAPDH. Independent experiments, n=3. P-values were calculated by two-way repeated-measures ANOVA followed by Tukey’s HSD post-hoc test. *P < 0.05. **Abbreviations: DTPA, diethylenetriamine pentaacetate; ALAS2, 5- aminolevulinic acid synthase2; ALAD, δ-aminolevulinic acid dehydratase; n.s., not significant. 58 Figure 2-15. The accumulation of total protoporphyrin IX is decreased by zinc restriction during erythroid terminal differentiation in G1E-ER4 cells. Total PPIX contents of non-differentiating and differentiating cells with or without DTPA treatment were measured at 48 hours of the differentiation in G1E-ER4 cells. The contents were normalized to the number of cells. Biological replicates, n=3. Values represent mean ± SD. P- values were calculated by one-way ANOVA followed by Tukey’s HSD post-hoc test. Different alphabet letters indicate statistically significant differences at P-value < 0.05. **Abbreviations: PPIX, protoporphyrin IX; β-Est, β-estradiol; DTPA, diethylenetriamine pentaacetate. 59 Figure 2-16. MEL-DS19 cells are suitable for analyzing the effects of zinc status on porphyrin metabolites. Fluorescence spectra of free PPIX and ZnPP (A) in MEL-DS19 cells treated with DMSO and (B) in G1E-ER4 cells with or without β-estradiol treatment are shown. **Abbreviations: DMSO, dimethylsulfoxide; β-Est, β-estradiol; PPIX, protoporphyrin IX; ZnPP, zinc protoporphyrin. 60 Figure 2-17. MEL-DS19 cells treated with DMSO undergo hemoglobinization. (A) Representative pellet images after 1.5% DSMO treatment into MEL-DS19 cells are shown. The samples were normalized to the same number of cells. (B) Heme contents and (C) relative total PPIX, (D) free PPIX, and (E) ZnPP in MEL-DS19 cells after DMSO treatment were measured. The contents were normalized to the number of cells. Biological replicates, n=3. Values represent mean ± SD. P-values were calculated by one-way ANOVA followed by Dunnett’s post hoc test. * P-value < 0.05 versus Day 0. **Abbreviations: MEL, MEL-DS19; DMSO, dimethylsulfoxide; ZnPP, zinc protoporphyrin. 61 Figure 2-18. Hemoglobin-related genes are elevated in MEL-DS19 treated with DMSO. Transcript abundance of Alas2, Hba-a1/2, and Hbb-b1 are presented three days after incubation with or without 1.5% DMSO. Genes of interest were normalized to Tbp. Biological replicates, n=3. Values represent mean ± SD. P-values were calculated by Student’s t-test. *P < 0.05. **Abbreviation: DMSO, dimethylsulfoxide; Alas2, aminolevulinic acid synthase-2; Hba-a1/2, hemoglobin alpha adult chain 1 and 2; Hbb-b1, hemoglobin beta adult chain 1. 62 Figure 2-19. Zinc deficiency immediately diminishes zinc protoporphyrin levels in MEL- DS19 cells treated with DMSO. (A) A scheme of experimental design of DTPA treatment in MEL-DS19 cells. (B) Representative pellet images of MEL-DS19 cells treated with DMSO for 5 days along different lengths of DTPA are shown. The contents were normalized to the same number of cells. (C) Representative fluorescence spectra for MEL-DS19 cells with DTPA and DMSO treatment containing free PPIX and ZnPP are shown. (D) Heme contents, relative (E) free PPIX and (F) ZnPP, and (G) ZnPP/Heme ratio in MEL-DS19 cells at five days of DMSO treatment were measured. No- DMSO was set at y = 1. The contents were normalized to the number of cells. Biological replicates, n=3. Values represent mean ± SD. P-values were calculated by one-way ANOVA followed by Dunnett’s post hoc test. *P < 0.05 versus the +DMSO −DTPA. **Abbreviations: DMSO, dimethylsulfoxide; DTPA, diethylenetriamine pentaacetate; ZnA, zinc-adequate; ZnD, zinc-deficient; PPIX, protoporphyrin IX; ZnPP, zinc protoporphyrin; d, days; n.s., not significant. 63 Table 2-3. Differentially expressed genes (DEGs) of GATA1-target genes by zinc restriction and the prediction on GATA1 activity identified by IPA. The transcriptome profile of zinc-restricted G1E-ER4 cells at 24-hour differentiation were analyzed. Differentially expressed genes (DEGs) of GATA1-target genes in the Ingenuity Pathway Analysis (IPA) dataset were identified by the criteria of |Fold change| > 2.0 and adjusted FDR P-value < 0.05. The observed fold changes of DEGs and the prediction on GATA1 direction based on the measurement direction and experimentally proved publications in the IPA database were shown in the Table. Observed fold changes ID Differentially expressed genes Prediction on GATA1 activity 85.849 ENSMUSG00000052217 HBE1 Activated 14.937 ENSMUSG00000020383 IL13 Activated 9.627 ENSMUSG00000029373 PF4 Activated 7.514 ENSMUSG00000020689 ITGB3 Activated 4.006 ENSMUSG00000020120 PLEK Activated 3.927 ENSMUSG00000050675 GP1BA Activated 3.328 ENSMUSG00000028163 NFKB1 Activated 3.209 ENSMUSG00000030054 GP9 Activated 3.093 ENSMUSG00000030067 Foxp1 Activated 2.74 ENSMUSG00000034117 PTGDR2 Activated 2.717 ENSMUSG00000025429 PSTPIP2 Activated 2.438 ENSMUSG00000000869 IL4 Activated 2.366 ENSMUSG00000003545 FOSB Activated 2.251 ENSMUSG00000026815 GFI1B Activated 2.249 ENSMUSG00000004266 PTPN6 Activated 2.126 ENSMUSG00000003429 RPS11 Activated 2.106 ENSMUSG00000004043 STAT5A Activated 2.031 ENSMUSG00000020160 Meis1 Activated -2.079 ENSMUSG00000027203 DUT Activated -2.295 ENSMUSG00000005732 RANBP1 Activated -2.368 ENSMUSG00000034041 LYL1 Activated -2.512 ENSMUSG00000022881 RFC4 Activated -3.008 ENSMUSG00000027342 PCNA Activated (Table continues on next page) 64 Observed fold change ID Differentially expressed genes Prediction on GATA1 activity 5.569 ENSMUSG00000027556 CA1 Inhibited 4.62 ENSMUSG00000005672 KIT Inhibited 3.753 ENSMUSG00000026069 IL1RL1 Inhibited 2.655 ENSMUSG00000024986 HHEX Inhibited 2.393 ENSMUSG00000025151 MAGED1 Inhibited 2.384 ENSMUSG00000025402 NAB2 Inhibited 2.342 ENSMUSG00000016477 E2F3 Inhibited 2.191 ENSMUSG00000022346 MYC Inhibited 2.136 ENSMUSG00000023990 TFEB Inhibited -2.035 ENSMUSG00000025400 TAC3 Inhibited -2.274 ENSMUSG00000031162 GATA1 Inhibited -2.515 ENSMUSG00000020472 ZNF496 Inhibited 22.219 ENSMUSG00000052187 HBG2 Affected 13.91 ENSMUSG00000074604 MGST2 Affected 7.441 ENSMUSG00000031465 ANGPT2 Affected 6.037 ENSMUSG00000022148 FYB1 Affected 5.186 ENSMUSG00000073940 HBB Affected 4.908 ENSMUSG00000019982 MYB Affected 4.676 ENSMUSG00000038418 EGR1 Affected 3.538 ENSMUSG00000060063 ALOX5AP Affected 2.967 ENSMUSG00000052920 PRKG1 Affected 2.726 ENSMUSG00000027646 SRC Affected 2.19 ENSMUSG00000001436 SLC19A1 Affected 2.166 ENSMUSG00000071636 RIMBP3 (includes others) Affected 2.136 ENSMUSG00000016087 FLI1 Affected -2.264 ENSMUSG00000023216 EPB42 Affected -2.349 ENSMUSG00000030086 CHCHD6 Affected 65 Discussion Approximately 1.1 billion people are at high risk of zinc deficiency worldwide (134). Anemia, defined as low red blood cell count or low hemoglobin content, is another major health concern, with an estimated global prevalence of 23% (65). The association between zinc deficiency and anemia has been reported in several clinical and laboratory studies. Zinc deficiency in humans was firstly documented in 1961, reporting a symptom similar to iron deficiency anemia (21,22). Plasma and serum zinc have been proposed as a predictor for hemoglobin concentrations and anemia risk in clinical studies, independent of iron status (2–5). These data emphasize the importance of zinc in understanding the etiology of anemia. Yet, how erythroid zinc deficiency affects the efficiency of red cell development remains unclear. Thus, the present study, employing G1E-ER4 and MEL-DS19 cells, determined the implications of zinc deficiency on heme biosynthesis and investigated mechanisms by which zinc deficiency affects hemoglobinization, particularly protoporphyrin synthesis, during terminal erythroid differentiation. Circulating mature erythrocytes contain approximately 10 times higher zinc concentration compared to plasma or serum (135), implying a physiological need for this nutrient by erythroid cells. A few animal studies have demonstrated the demand for zinc in red blood cell production and functioning. In the bone marrow of rats, zinc assimilation was observed in the condition of increased erythropoietic demand due to hemolytic anemia (6). Additionally, redistribution of zinc was reported in anemic mice and rats, leading to elevated bone marrow and erythrocyte zinc (68). More recently, 66 upregulation of the zinc importer ZIP10 and downregulation of the exporter ZnT1 in the plasma membrane fraction of mature red blood cells by dietary zinc restriction have been demonstrated in mice (85). These observations are in agreement with our finding in the current study that identifies a physiological demand for zinc by erythroid cells, particularly during their development into mature red cells. Zinc deficiency has been associated with red blood cell metabolisms. Dietary zinc restriction shortened the lifespan of erythrocytes due to membrane fragility in rats and pigs (35,36). In addition, a zinc-deficient diet led to decreased erythropoietin synthesis in rats (71). The present study mechanistically demonstrates impaired hemoglobinization by restriction of the cellular zinc supply to developing erythroid progenitors. Impaired protoporphyrin synthesis but not inadequate cellular iron, of which both can result in heme deficiency, were identified as results of zinc restriction. Zinc is a cofactor for ALAD, which is involved in PPIX synthesis by mediating the conversion of δ-ALA to PBG (10,16). In rats, lower erythroid ALAD activity has been observed by consumption of a zinc-deficient diet (136). Moreover, ALAD activity in red blood cells has been shown as responsive to the intake of zinc supplementation in humans (137). Our data demonstrate that zinc deficiency reduces not only the activity of ALAD but also represses its expression. The regulatory mechanism by which zinc controls the transcript abundance warrants further investigation. The present study indicates that the impact of zinc restriction on erythroid heme deficiency can differ by the stage of erythroid development. Erythroid zinc in early differentiation might be important as a structural component of zinc-finger transcription factors to terminal erythroid differentiation and cell survival (8,76,138). Recently, an 67 expansion in the cellular zinc pool in the early stage of differentiation has been demonstrated as essential for the survival of erythroid progenitors during differentiation, further supporting the need for zinc in the early stages of terminal erythroid differentiation (8). However, in the same study, the intracellular zinc pool of developing erythroid progenitors was shown to shrink once the cells reached a particular stage of terminal development (8). Additionally, excessive zinc ion has been shown to interfere with heme production by competing with iron as a substrate for ferrochelatase to form zinc protoporphyrin (115), which could be toxic in excess (139). Our studies, showing the requirements of zinc early in development but not in later stages, are in agreement with these findings, suggesting a shift in the physiological roles of zinc during terminal erythroid differentiation. A reliable biomarker or status indicator for zinc deficiency is lacking. In a clinical setting, hemoglobin content and erythrocyte morphology or size are commonly assessed as indices for nutrient status, e.g., iron or vitamin B12 deficiency which causes microcytic and macrocytic anemia, respectively (140,141). Additionally, iron deficiency anemia can be defined by a rise in ZnPP-to-heme ratio (115,141), which is caused by the relative increase in zinc as a substrate for ferrochelatase when cells become iron- restricted. Here, we demonstrate that ZnPP contents drop by zinc restriction in cells, resulting in a decline in the ZnPP-to-heme ratio. Thus, a low zinc protoporphyrin/heme ratio could function as a potential diagnostic approach to differentiate iron deficiency anemia from anemia caused by zinc deficiency. Zinc functions as a structural component of GATA1, an erythroid-specific transcription factor (74). G1E-ER4 cells are engineered to activate inducible GATA1 in 68 the presence of β-estradiol (142). Upon the terminal erythroid differentiation, GATA1 upregulates Tfrc, an iron-importing gene, and Alas2, encoding the rate-limiting enzyme involved in protoporphyrin synthesis (143). This indicates that zinc-deficient G1E-ER4 cells differentiate with normal Alas2 and Tfrc transcript responses at 24 hours. Moreover, the bioinformatic analyses of the transcriptome of G1E-ER4 cells differentiating under zinc restriction identified higher activities of GATA1 rather than a decline in the transcription factor due to zinc restriction. The hierarchical clustering suggests that differentiating erythroid progenitors with zinc restriction feature unique molecular profiles compared to those that were normally differentiating and that of the cells that had their GATA1 kept inactive. Collectively, the impaired development of erythroid progenitors by zinc restriction was not merely due to impaired induction of differentiation, i.e., lower GATA1 activity, but rather caused by impairments in metabolic processes mediating erythroid development thereafter. Anemia is defined as a condition resulting in fewer red blood cells in circulation and lower hemoglobin content in the blood (66). Thus, either lower numbers of total red cells or a decline in mean cell hemoglobin contents may contribute to the development of anemia. Cellular zinc import, particularly by ZIP8, has been identified essential for the survival of erythroid progenitors during their maturation (8). In other words, a loss in developing erythroid progenitors by impaired cellular zinc acquisition may lead to lower erythrocyte production via erythropoiesis. The current studies identified impaired hemoglobinization prior to any losses in cell count, viability, or cell size caused by zinc restriction. Thus, zinc deficiency may contribute to the development of anemia by either reducing the number of newly produced red blood cells via erythropoiesis or by 69 impairing heme production by each erythroid progenitor. Whether zinc status has a larger impact on erythroid cell number or cellular heme content in live animals or humans warrants further investigation via in vivo experimentation. In nonerythroid cells, zinc has been associated with regulated cell death, particularly apoptosis. Zinc functions as a regulatory factor for p53 and caspase 3, and cellular zinc depletion can induce apoptosis of leukemia cells and macrophages (126,127,144). Additionally, zinc has been characterized to have protective roles against potentially cytotoxic oxidative stress (145). In rats, a zinc-deficient diet impaired the antioxidant defense system and increased oxidative stress in the erythrocytes (69,70). Erythrocyte Cu/Zn SOD (SOD1), a zinc-containing anti-oxidative enzyme, is involved in the defense system against elevated oxidative stress (13). Developing erythroid progenitors experience a massive expansion in their cellular iron pool during terminal erythroid differentiation. While iron is integral to heme biosynthesis, it may impose excess oxidative stress on cells by producing reactive oxygen species such as hydroxyl radicals (146). Of relevance is a recently discovered novel form of regulated cell death, ferroptosis, which is induced by the release of redox-active iron atoms into the cytosol (121,146). Bioinformatic comparisons of the transcriptome profiles of zinc-adequate and zinc-restricted erythroid progenitors allowed us to identify the enrichment of gene responses associated with ferroptosis (e.g. Hmox1, SLC7A11, Abca1, Prdx6, Hmgcr, Fdft1, and Fancd2). Moreover, extensive zinc restriction or acute depletion using a cell- permeable chelator (Figure 3-2) caused losses in the viability of erythroid cells undergoing differentiation. The findings from the current study may imply a role of zinc in defense against iron-induced oxidative stress, which may eventually lead to 70 ferroptosis. Of particular relevance are the increase and decrease in Tfrc mRNA and IRP2 proteins, respectively, which are indicative of an expansion in the cytosolic labile iron pool (147), by extensive zinc restriction in developing G1E-ER4 cells. Additionally, a recently identified genetic factor for defense against ferroptosis, the Prdx6 gene, was identified responsive to zinc restriction. Collectively, these findings support the role of zinc in the regulation of the antioxidant system and ferroptosis in erythroid cells, naturally programmed to experience high levels of intracellular iron and thus potential oxidative stress during their development (148). In this chapter, our findings mechanistically demonstrate the requirement of adequate zinc supply to developing erythroid cells for not only supporting their proliferation and survival (8) but also for facilitating their core metabolic process, heme production. To our knowledge, these are the first studies to define the essentiality of zinc for normal erythroid hemoglobinization. Based on these findings, future in vivo animal or clinical studies should aim to determine the essentiality of dietary zinc intake for normal erythroid maturation and prevention or treatment of anemia. Additionally, anemia remains a primary nutrition disorder worldwide (65), and many cases remain ineffectively treated due to their unclear etiology. The validation of ZnPP-to-heme ratio as a relevant diagnostic tool for zinc deficiency anemia will permit correction of anemia cases attributable to zinc deficiency. 71 Chapter 3 . Zinc transporter ZIP10 is required for heme biosynthesis and cell survival during red cell development 72 Introduction Cellular zinc homeostasis is achieved via the cell-type-specific and gene-specific regulatory mechanisms of zinc transporters (10). Zinc transporters, comprising 10 members of ZnT protein and 14 members of ZIP protein, mediate the mobilization of zinc across plasma and intracellular membranes (35). ZnT encoded by the SLC30A gene family facilitates zinc efflux from the cytosol to either the extracellular space or the lumens of intracellular compartments. Conversely, ZIP protein encoded by the SLC39A gene family mediates the zinc influx from either the extracellular space or intracellular compartments into the cytosol (10). A few studies have identified the expression and physiological significance of zinc transporters in erythroid cells. ZIP8 deficiency in mice leads to severe anemia due to dysregulated hematopoiesis (86). Moreover, differential regulations of zinc transporters have been proposed as mechanisms mediating stage-dependent changes in the cellular zinc contents during terminal erythroid differentiation (8,85). Zinc has known anti- apoptotic roles (126,127) while may interfere with the last step of heme biosynthesis, which is to incorporate iron into PPIX molecules (115). The upregulation of zinc importer Zip8 preceding the activation of exporter ZnT1 (8,85) allows an adequate supply of zinc required for cellular survival early without interfering with heme biosynthesis later during erythroid differentiation (8). ZIP8, ZIP10, and ZnT1 proteins are present in the plasma membrane of mature red blood cells in mice (85). Among these, ZIP10 and ZnT1 have been characterized as zinc-regulated transporters (85). In replete zinc status, Zip10 transcription becomes 73 repressed and ZnT1 is activated, both via the binding of MTF-1 to the MRE downstream and upstream of its transcription start site, respectively (52). In mature erythrocytes from mice, ZIP10 and ZnT1 proteins located on the plasma membrane were upregulated and repressed by a zinc-depleted diet, respectively (85). This implies the physiological need for tight regulation and maintenance of cellular zinc homeostasis in erythroid cells. The studies of this chapter identify Zip10 as the zinc transporter gene most responsive to cellular zinc depletion in developing erythroid progenitors and its role in compensatory zinc import during mild cellular zinc restriction. The findings not only support the importance of zinc in erythroid development and hemoglobinization as described in the previous chapter but also introduce a novel nutrient-gene interaction with pathophysiological implications for hematological diseases, including anemia Materials and Methods Cell line and treatments G1E-ER4 cell line and DTPA were employed as described in Chapter 2. Cellular zinc deficiency was achieved by zinc-chelating agents, N,N,N′,N′-tetrakis(2- pyridylmethyl)-ethylenediamine (TPEN) and DTPA. TPEN is a cell-permeable zinc- chelator for severe and instant zinc depletion, whereas DTPA is an impermeable zinc- chelator to limit the extracellular availability of zinc (118). The final concentrations of TPEN (10 μM) and DTPA (10-50 μM) were determined based on the previous studies (52,118). Heme production was inhibited by adding an iron chelator, deferoxamine (DFO), or an ALAD inhibitor, succinylacetone (SA), to the culture medium (149). DFO 74 was added at a final concentration of 50 μM for 48 hours, and SA was treated at a concentration of 0.2 mM twice, at the beginning and 24-hour of differentiation. Gene silencing by short interfering RNA (siRNA) For the gene silencing, G1E-ER4 cells were transfected with siRNA twice via a Nucleofector II device (Lonza), with an interval of 24 hours. Six million G1E-ER4 cells were mixed with 480 pmoles of siRNA in 100 μL of Nucleofector Solution R with supplement (Lonza) and electroporated using a Nucleofector II device. Silencer Select siRNA (4390815, Ambion) was utilized for ZIP10 RNAi. Negative control transfection was carried out with non-targeting siRNA (4390846, Invitrogen). The transfected cells at a density of 2 x 105 cells/mL were induced the differentiation immediately after the second transfection by adding 100 nM β-Est for up to 48 hours. Major minerals and trace elements analyses ICP-MS of transfected cells was conducted as previously described in Chapter 2. Quantitative RT-PCR and Western blot qPCR and Western blot were carried out as previously described in Chapter 2. Immunoblotting was conducted by primary incubations with rabbit anti-ZIP10 (1:1,000; Invitrogen, PA5-21064) primary antibody and primary antibodies listed in Chapter 2. Zinc transporter primers for qPCR amplification are listed in Table 3-1 and Table 3-2. Heme assays Cellular heme contents were measured as previously described in Chapter 2. Transcriptome and bioinformatic analyses RNA-seq and post-hoc bioinformatic analyses of normalized sequencing data were conducted as previously described in Chapter 2. Venn diagrams of DEGs by DTPA 75 treatment and ZIP10-depletion were drawn using VENNY 2.1 (https://bioinfogp.cnb.csic.es/tools/venny/). Statistics All in vitro experiments include at least three biological replicates or independent experiments. Data were presented as mean ± standard deviation. Student’s t-test, one-way ANOVA, or two-way ANOVA followed by Dunnett’s or Tukey's HSD post-hoc test was conducted to define significant differences. A P-value < 0.05 was considered statistically significant. 76 Table 3-1. ZnT primers for SYBR Green-based qPCR analyses. Gene of Interest Primer direction Primer sequence Slc30a1 (ZnT1) Forward 5' - TCCGACTCCTTCCACATGCT - 3' Reverse 5' - CCGAACGTGTTCTTCTGCGT - 3' Slc30a2 (ZnT2) Forward 5' - TCGAGCCGAGATCCTTGGAG - 3' Reverse 5' - GCAGCCCGAAGTGATCAACA - 3' Slc30a3 (ZnT3) Forward 5' - ACCCGTAAGGGACACCTTGT - 3' Reverse 5' - AGCCGTGGAGTCAATAGCCA - 3' Slc30a4 (ZnT4) Forward 5' - GGCTTTCACAACGTTTCGCAT - 3' Reverse 5' - AATGGCTTGGTACACCTTCCAG - 3' Slc30a5 (ZnT5) Forward 5' - TGCTGACACCAGTTTCCGTC - 3' Reverse 5' - CATGGTGTGAATGGCCGTGA - 3' Slc30a6 (ZnT6) Forward 5' - TTTGGCTCATTGGCTGGGTC - 3' Reverse 5' - TGTTCGTTCGCATCTCGTCG - 3' Slc30a7 (ZnT7) Forward 5' - ACTTCTAGAGGGACGGAGACC - 3' Reverse 5' - GACAGGATGGACCTAAACCAGC - 3' Slc30a8 (ZnT8) Forward 5' - TATCGAGCAGAGATCCTCGGTG - 3' Reverse 5' - GCTCACAGGCAAGGTACAGC - 3' Slc30a9 (ZnT9) Forward 5' - GTGCAGCGGCTTACTGAACT - 3' Reverse 5' - GGATTGCCCTTACTGACGGG - 3' Slc30a10 (ZnT10) Forward 5' - ACATGGAAGAGCTGATGAGCCA - 3' Reverse 5' - GCTGGCATCCTGGTATTCCG - 3' 77 Table 3-2. Zip primers for SYBR Green-based qPCR analyses Gene of Interest Primer direction Primer sequence Slc39a1 (Zip1) Forward 5' - CCTGCATGTGACGCTTCAGT - 3' Reverse 5' - CCAGCGTGATCTGCTCCATC - 3' Slc39a2 (Zip2) Forward 5' - TGGTTCTCACACTGGGCTGT - 3' Reverse 5' - CTGTGGTGATGACCTGTAGCTG - 3' Slc39a3 (Zip3) Forward 5' - GCACCAAGCCTCAAGCTTCT - 3' Reverse 5' - ACACAGGGCCAAAGCACAAA - 3' Slc39a4 (Zip4) Forward 5' - GCACAGCCACCCACTACATC - 3' Reverse 5' - AGTCCCAGCACCTTGGGTAT - 3' Slc39a5 (Zip5) Forward 5' - TGACAGCCGTGTTTGCATCA - 3' Reverse 5' - TGAAGCAGGGCAGACAGTACA - 3' Slc39a6 (Zip6) Forward 5' - TCTCTGCCCAGCCATCATCA - 3' Reverse 5' - GCCACCAAGCCAGGCTATTT - 3' Slc39a7 (Zip7) Forward 5' - CGCATGCCTTGGAACCTCATT - 3' Reverse 5' - CCACGAGGAAGGCGACAATC - 3' Slc39a8 (Zip8) Forward 5' - AGCTGCACTTCAACCAGTGTT - 3' Reverse 5' - TCCTCGCAGGGATGGAAGTT - 3' Slc39a9 (Zip9) Forward 5' - TGTGCATTCCAGTGATGATCCA - 3' Reverse 5' - CTGCAGCATGGACGACTAGC - 3' Slc39a10 (Zip10) Forward 5' - ACCGCCAGCATGAATGTTTGA - 3' Reverse 5' - ATGCAGGGCAAAGGTACGTG - 3' Slc39a11 (Zip11) Forward 5' - GTGCCAGGAATCTGGCCATTG - 3' Reverse 5' - CCGCTCAGCTGTCCATACCA - 3' Slc39a12 (Zip12) Forward 5' - TTGGAGAGGACTGGGTGTCC - 3' Reverse 5' - GCTTCCAAACACAGCTTGCAG - 3' Slc39a13 (Zip13) Forward 5' - TACAGTCAGAAGCCGGAGCC - 3' Reverse 5' - CCAGGAGTCCACCTAAGGCA - 3' Slc39a14 (Zip14) Forward 5' - CTCCATGTCTGTGCAGGACC - 3' Reverse 5' - GCCAGTAGCAAGCACTCTGG - 3' 78 Results Zip10 is the most responsive gene to zinc restriction among 24 zinc transporters Previous studies identified ZIP8, ZIP10, and ZnT1 as zinc transporters expressed by developing and mature erythroid cells (8,85). Among these, ZIP10 and ZnT1 of mature erythrocytes presented differential expression by dietary zinc deficiency in mice (85). However, whether these transporters or other zinc transporter genes respond to cellular zinc status in erythroid progenitors remains unknown and thus was tested. Fold changes of each zinc transporter gene by zinc restriction were assessed through the previously described RNA-seq dataset of DTPA-treated and control G1E-ER4 cells (Figure 3-1 A). Zip10 transcript was identified to be most responsive to zinc restriction during erythroid development. This finding from RNA-seq-based transcriptome analysis was further validated by individual qPCR assays for each of all 24 zinc transporter gene transcripts (Figure 3-1 B). TPEN is a cell-permeable zinc chelator that can acutely deplete the cellular contents of bioavailable zinc. To further determine the acute and direct effects of intracellular zinc depletion on the regulation of zinc transporters, differentiating G1E- ER4 cells were treated with TPEN. Since TPEN treatment for 24 hours in proliferating and developing G1E-ER4 cells led to acute losses in viable cells (Figure 3-2), shorter treatments (1 or 3 hours) were used for the gene expression analyses. As by DTPA, a prominent upregulation of Zip10 transcript was produced by 3 hours of TPEN treatment (Figure 3-3). The response of Zip10 transcript to intracellular zinc depletion occurred as fast as one hour after TPEN treatment and prior to any transcript changes in globin genes, 79 Hba-a1/2 and Hbb-b1, and the transcription factor, Gata1 (Figure 3-4 A). The Zip10 response to zinc depletion was also observed in proliferating cells, indicating that the acute response is primarily due to zinc depletion rather than changes in differentiation status or efficiency (Figure 3-4 B). Zip10 is suppressed during terminal erythroid differentiation but is highly upregulated by zinc deficiency. A previous study demonstrated that ZnT1, ZIP8, and ZIP10 exist on the mature red cell membrane in mice (85), and their temporal regulation during terminal erythroid differentiation has been demonstrated using erythroid splenocytes isolated from mice experiencing hemolytic anemia (85). G1E-ER4 cells were employed as an alternative model of erythroid differentiation, and regulations of ZnT1, Zip8, Zip10, and metallothionein genes were measured (Figure 3-5). While increases in transcripts of Mt1, ZnT1, and Zip8 were observed by the differentiation of G1E-ER4 cells, Zip10 mRNA levels were suppressed as the cells initiated differentiation and stayed low throughout the development. In the previous zinc transporter screening experiments, Zip10 was identified as a transporter gene highly responsive to zinc restriction in mature erythrocytes (85) and G1E-ER4 erythroid progenitors, which have gone through 24 hours of differentiation (Figure 3-1). Thus, whether cellular zinc status affects the temporal gene expression of Zip10 early in and throughout differentiation was tested by a time-course study. Zip10 transcript abundances remained steadily high by zinc restriction (Figure 3-6). The upregulation of ZIP10 expression by zinc restriction was confirmed at the protein level (Figure 3-7), and it is in agreement with the ZIP10 response to dietary zinc restriction in 80 mature erythrocytes (85). These findings indicate that the regulation of ZIP10 by zinc is not limited to later stages of erythroid differentiation and persists throughout development. As shown earlier, zinc restriction causes heme deficiency in differentiating erythroid progenitors. Heme may function as a regulator of gene expressions (128). Thus, we tested whether cellular heme status may contribute to the upregulation of Zip10 expression by zinc restriction (Figure 3-1) and its repression by differentiation seen earlier (Figure 3-5). For this, heme deficiency in G1E-ER4 cells was induced by an iron chelator, DFO, and a known ALAD inhibitor, succinylacetone (SA) (Figure 3-8). After 48 hours of differentiation, the treatments by DTPA, DFO, and SA to G1E-ER4 cells produced a comparable magnitude of the decline in cellular heme contents. However, only DTPA treatment affected Mt1 transcript abundance and Zip10 response (Figure 3-9). The specificity of Zip10 response to zinc was further confirmed by the reversal of DTPA- induced repression by the addition of zinc matched by molarity (Figure 3-10). Accordingly, these findings demonstrate that developing erythroid progenitors control their levels of Zip10 expression in response to cellular zinc availability in a manner independent to heme status. ZIP10 functions an essential role in hemoglobinization under zinc restriction during red cell development. To determine the functional importance of ZIP10 in erythroid progenitors, the loss-of-function approach employing siRNA-mediated gene silencing was employed. Effective repression of Zip10 expression was confirmed by a 64.6% decrease in Zip10 transcript abundance by the transfection of Zip10-specific siRNA transfection versus the 81 non-specific control siRNA transfection (Figure 3-11). Zip10 silencing alone did not produce any changes in molecular indices of heme biosynthesis in G1E-ER4 cells differentiating in the control medium providing an adequate level of zinc to the cells (Figure 3-12). The higher levels of ZIP10 expression by zinc restriction would imply a role of the protein for improved zinc import efficiency when the extracellular availability of zinc becomes limited. To test this, a cell culture condition where zinc is low but still enough for normal heme production by differentiating G1E-ER4 was defined. Previous data showed that the impairment of hemoglobinization occurs by DTPA at 50 μM but not 10 or 25 μM (Figure 2-2). To determine if the lower doses of DTPA still produce cellular zinc restriction despite the lack of impaired heme synthesis and a compensatory upregulation of ZIP10, Mt1 and Zip10 mRNA levels in G1E-ER4 cells treated with different DTPA doses were measured. Mt1 transcript abundance significantly decreased regardless of the dose of DTPA treatments, while Zip10 mRNA responded to DTPA in a dose-dependent manner (10, 25, and 50 μM) (Figure 3-13). We selected the 25 μM DTPA treatment for our subsequent loss-of-function studies on ZIP10, as a condition producing the highest degree of cellular zinc restriction, yet mild enough to prevent any impact on heme biosynthesis (Figure 2-2). To assess the role of ZIP10 during red cell development under zinc restriction, heme contents were measured. Visual inspection of cell pellet exhibited less red coloration in ZIP10-deficient G1E-ER4 cells under mild zinc restriction (by 25 μM DTPA) (Figure 3-14). As previously shown, 25 μM DTPA or ZIP10 deficiency alone had no effects on the appearance of cell pellets. Measures of cellular heme content were in 82 agreement with the intensity of red coloration of each cell pellet. Additionally, α-globin mRNA and protein abundance were repressed by the mild DTPA treatment only when cells were depleted of ZIP10 (Figure 3-15). These data demonstrate that ZIP10 becomes essential for normal erythroid development and heme production when erythroid cells develop under zinc restriction. ZIP10 deficiency impairs zinc homeostasis during red cell development under zinc restriction. To elucidate ZIP10 function on zinc status, the total cellular zinc contents were analyzed via ICP-MS-based metal quantitation in cells differentiating with or without ZIP10 silencing. At 24 hours of differentiation, loss in ZIP10 alone had no effect on cellular zinc contents (Figure 3-16 A). However, ZIP10 depletion resulted in lower cellular zinc accumulation when zinc supply was restricted by co-treatment of DTPA at 25 μM (Figure 3-16 B). Mt1 transcript expression did not reflect a decrease in cellular zinc contents measured at the 24-hour time-point of ZIP10 depletion and zinc restriction (Figure 3-16 C). Thus, we measured Mt1 mRNA abundance at earlier time points. An adequate supply of zinc in early differentiation is important for heme biosynthesis (Figure 2-7). Depletion of ZIP10 under zinc restriction resulted in lower Mt1 mRNA levels early as 8 hours of differentiation (Figure 3-17). These support the role of ZIP10 in the maintenance of cellular zinc homeostasis early in terminal erythroid differentiation, particularly when the cellular zinc supply becomes restricted. ZIP10 depletion under zinc restriction is associated with ferroptosis cell death. To gain an unbiased and comprehensive understanding of zinc-restricted and ZIP10-depleted erythroid development, we performed a transcriptome profiling using 83 RNA-seq. The numbers of DEGs were 342 by zinc restriction (178 upregulated and 162 downregulated), 284 (85 upregulated and 199 downregulated) by ZIP10 depletion, and 2094 (1327 upregulated and 767 downregulated) by co-treatment, when a threshold for significance at |Fold change| > 2.0 and FDR p-value < 0.05 was applied (Figure 3-18). The number of DEGs produced by each treatment and shared by more than two treatments are shown in Figure 3-18. The numbers of unique DEGs were 73, 62, and 1706 by zinc restriction, ZIP10 depletion, and co-treatment, respectively. Comparison analysis by IPA revealed 84 canonical pathways significantly enriched by the co-treatment of ZIP10 depletion and zinc restriction at of P-value < 0.01 and Z-score >1 (Filter: All species, All cell type) (Table 3-3). Notably, the ferroptosis signaling pathway was identified as a significantly activated canonical pathway (Z-score = +1.5) by the combination of ZIP10 depletion and zinc restriction. Similar to the cell loss by the higher dose of DTPA treatment (Figure 2-4), the combination of ZIP10 depletion and mild zinc restriction reduced cell viability (Figure 3-19). This observation further supports the well-characterized roles of zinc in the regulation of cell death (126,127,144). The DEGs related to the ferroptosis signaling pathway listed in IPA are summarized in Table 3-4 with their corresponding fold-changes by 2.0 (FDR p-value < 0.05). 84 Figure 3-1. Zip10 is the most responsive gene to zinc restriction among 24 zinc transporters in developing erythroid progenitors. (A) A heat map shows transcript fold changes of TPM by 50 μM DTPA treatment at 24 hours of differentiation from RNA-seq data in log2 scale. (B) qPCR confirmation was conducted from the same biological samples. Biological replicates, n=3. Genes of interest were normalized to Tbp. Zinc transporter genes with undetectable expression are not shown. **Abbreviations: TPM, transcript per million; DTPA, diethylenetriamine pentaacetate; ZnA, zinc-adequate; FC, fold change. 85 Figure 3-2. TPEN treatment leads to significant cell loss in proliferating and developing G1E-ER4 cells. (A) A experimental design of 10 μM TPEN treatment in G1E-ER4 cells. The black bar indicates the TPEN-treated period. (B) Cell viability of TPEN-treated G1E-ER4 cells with or without β- Est. Biological replicates, n=4. Values represent mean ±SD. P-values were calculated by one-way ANOVA followed by Dunnett’s post-hoc test. *P < 0.05 versus 0 h-TPEN treatment. (C) Representative cell microscope images of G1E cells treated with or without 10 μM TPEN and treated with or without β-Est. **Abbreviations: β-Est, β-estradiol; TPEN, N,N,N′,N′-tetrakis(2- pyridylmethyl)-ethylenediamine. 86 Figure 3-3. Transcript responses of zinc transporter genes to TPEN treatment in differentiating G1E-ER4 Cells. Differentiating G1E-ER4 cells were treated with 10 μM of TPEN 3 hours prior to the harvesting at 24-hour of the differentiation. Genes of interest were normalized to Tbp. Biological replicates, n=3. **Abbreviations: TPEN, N,N,N′,N′-tetrakis(2-pyridylmethyl)-ethylenediamine; Mt, metallothionein; ZnA, zinc-adequate. 87 Figure 3-4. Zip10 transcript immediately and directly responds to zinc deficiency. Zinc transporters respond to 10 μM of TPEN treatment for 1 or 3 hours (A) at 24 hours of differentiation and (B) during proliferation in G1E-ER4 cells. Genes of interest were normalized to Tbp. Independent experiments, n=3. Values represent mean ± SD. P-values were calculated by one-way ANOVA followed by Dunnett’s post-hoc test. *P < 0.05 versus ZnA. **Abbreviations: TPEN, N,N,N′,N′-tetrakis(2-pyridylmethyl)-ethylenediamine; ZnA, zinc-adequate. 88 Figure 3-5. Temporal patterns of transcripts of zinc transporters and metallothionein during terminal erythroid differentiation. Transcript abundance of (A) Zip10, (B) Mt1, (C) Zip8, and (D) ZnT1 is presented upon β-Est treatment time. Genes of interest were normalized to Tbp. Independent experiments, n=3. Values represent mean ± SD. *P < 0.05 compared to the same time point by two-way ANOVA followed by Tukey’s HSD post-hoc test. **Abbreviations: β-Est, β-estradiol; DTPA, diethylenetriamine pentaacetate. 89 Figure 3-6. Zinc restriction immediately increases Zip10 mRNA levels during the terminal erythroid differentiation. Temporal trends of (A) Mt1 and (B) Zip10 mRNA levels are presented upon β-Est treatment time. Genes of interest were normalized to Actb. Biological replicates, n=3. Values represent mean ± SD. *P < 0.05 by two-way ANOVA followed by Tukey’s HSD post-hoc test. **Abbreviations: β- Est, β-estradiol; DTPA, diethylenetriamine pentaacetate. 90 Figure 3-7. Zinc restriction increases ZIP10 protein abundance during red cell development. (A) Western blot image of ZIP10 treated with or without DTPA at 0, 24, and 48 hours of differentiation. (B) Protein abundance of ZIP10 by DTPA was assessed at 0 and 24 hours of differentiation. GAPDH was detected for normalization. Independent experiments, n=3. Values represent mean ± SD. P-values were calculated by one-way repeated-measures ANOVA followed by Tukey’s HSD post-hoc test. *P < 0.05. **Abbreviations: β-Est, β-estradiol; DTPA, diethylenetriamine pentaacetate. 91 Figure 3-8. Heme deficiency is caused by different factors in developing G1E-ER4 cells. (A) A simple scheme of heme biosynthesis with heme inhibitors. (B) Representative pellet images of the developing G1E-ER4 cells at 48 hours of differentiation. The cells were incubated with the heme inhibitors including DTPA, DFO, and SA. (C) Heme contents of the cells treated with DTPA, DFO, and SA were measured at 48 hours of the development. Independent experiments, n=3. Values represent mean ± SD. P-values were calculated by one-way ANOVA followed by Tukey’s HSD post-hoc test. *P < 0.05. **Abbreviations: CON, control; DFO, deferoxamine; SA, succinylacetone; β-Est, β-estradiol; DTPA, diethylenetriamine pentaacetate. 92 Figure 3-9. Zip10 only responds to zinc restriction, not to heme deficiency caused by iron or protoporphyrin deficiency during red cell development. Transcript abundance of Hba-a1/2, Mt1 and Zip10 was measured at 24 hours of the G1E-ER4 differentiation. Genes of interest were normalized to Tbp. Biological replicates, n=3. Values represent mean ± SD. P-values were calculated by one-way ANOVA followed by Tukey’s HSD post-hoc test. *P < 0.05 versus CON. **Abbreviations: β-Est, β-estradiol; DTPA, CON, control; DTPA, diethylenetriamine pentaacetate; DFO, deferoxamine; SA, succinylacetone. 93 Figure 3-10. Zip10 is responded to zinc restriction by DTPA and fully reversed by the addition of equimolar zinc. Transcript abundance of Mt1 and Zip10 was measured at 24 hours of the G1E-ER4 differentiation after adding back equimolar zinc (50 μM) to DTPA-treated cells. Genes of interest were normalized to Tbp. Biological replicates, n=3. Values represent mean ± SD. P-values were calculated by one-way ANOVA followed by Tukey’s HSD post-hoc test. *P < 0.05. **Abbreviations: CON, control; DTPA, diethylenetriamine pentaacetate; n.s., not significant. 94 Figure 3-11. Gene silencing approach using ZIP10 siRNA suppresses Zip10 mRNA abundance. (A) A scheme of siRNA experiment. (B) Transcript abundance of Zip10 was measured at 24 hours of the double-transfected G1E-ER4 differentiation using siRNA. Genes of interest were normalized to Tbp. Biological replicates, n=3. P-values were calculated by Student’s t-test. *P < 0.05. Values represent mean ± SD. **Abbreviations: β-Est, β-estradiol; siCON, Control siRNA- transfected cells; siZIP10, ZIP10 siRNA-transfected cells. 95 Figure 3-12. ZIP10 depletion alone did not produce changes in molecular indices of heme biosynthesis during red cell development. (A) Transcript abundance of Hba-a1/2 was measured at 24 hours of the double-transfected G1E- ER4 differentiation. Genes of interest were normalized to Tbp. Biological replicates, n=3. (B) Heme contents of the transfected cells were measured at 48 hours of the development. Independent experiments, n=3. Values represent mean ± SD. *P < 0.05 by Student’s t-test. **Abbreviations: siCON, Control siRNA-transfected cells; siZIP10, ZIP10 siRNA-transfected cells; n.s., not significant. 96 Figure 3-13. Zip10 responds to zinc restriction by DTPA in a dose-dependent manner. Transcript abundance of (A) Mt1 and (B) Zip10 was measured at 24 hours of differentiation with different doses (10, 25, and 50 μM) of DTPA. Genes of interest were normalized to Tbp. Biological replicates, n=3. Values represent mean ±SD. P-values were calculated by one-way ANOVA followed by Tukey’s HSD post-hoc test. *P < 0.05. Different alphabet letters indicate statistically significant differences at P-value <0.05. **Abbreviation: DTPA, diethylenetriamine pentaacetate. 97 Figure 3-14. ZIP10 is required under zinc restriction for heme production during red cell development. (A) Representative pellet images of the transfected cells at 48 hours of the differentiation treated with DTPA (25 μM) are shown. (B) Heme contents of the transfected cells were treated with 25 μM DTPA at 48 hours of the development. Independent experiments, n=3. Values represent mean ±SD. P-values were calculated by two-way ANOVA followed by Tukey’s HSD post-hoc test. Data connected by lines are significantly different at P < 0.05. **Abbreviations: β-Est, β- estradiol; siCON, Control siRNA-transfected cells; siZIP10, ZIP10 siRNA-transfected cells; DTPA, diethylenetriamine pentaacetate. 98 Figure 3-15. ZIP10 depletion results in the impairments of alpha-globin protein and transcript under zinc restriction during red cell development. (A) Transcript abundance of Hba-a1/2 at 24 hours of differentiation. Gene of interest was normalized to Tbp. Biological replicates, n=4. (B) Representative western image of α-Globin protein expressions by DTPA treatment and siZIP10 employment. (C) Relative protein expressions of α-Globin were measured at 24 and 48 hours. GAPDH was detected for normalization. Independent expriments, n=3. Values represent mean ±SD. P-values were calculated by two-way ANOVA followed by Tukey’s HSD post-hoc test. Data connected by lines are significantly different at P < 0.05. **Abbreviations: β-Est, β-estradiol; siCON, Control siRNA-transfected cells; siZIP10, ZIP10 siRNA-transfected cells; DTPA, diethylenetriamine pentaacetate. 99 Figure 3-16. ZIP10 depletion reduces cellular mineral contents under zinc restriction during red cell development. Cellular mineral contents were measured in developing G1E-ER4 cells (A) without or (B) with 25 μM of DTPA at 24-hour differentiation via ICP-MS-based metal quantitation. The results were normalized to gram protein. Phosphorous contents were measured as an internal control. Biological replicates, n=4. Biological replicates, n=3. P-values are calculated by Student’s t-test. (C) Transcript abundance of Mt1 and Zip10 was measured at 24-hour differentiation with or without 25 μM DTPA treatment. Genes of interest were normalized to Tbp. Biological replicates, n=3. Values represent mean ± SD. P-values are calculated by two-way ANOVA followed by Tukey’s HSD post hoc test. Different alphabet letters indicate statistically significant differences at P-value <0.05. **Abbreviations: siCON, Control siRNA-transfected cells; siZIP10, ZIP10 siRNA-transfected cells; DTPA, diethylenetriamine pentaacetate; Zn, zinc; P, phosphorous; Fe, iron. 100 Figure 3-17. ZIP10 depletion affects Mt1 expressions from the early terminal erythroid development under zinc restriction. (A) A sheme of siRNA experiment. Transcript abundance of (B) Zip10 and (C) Mt1 was measured throughout the early development with or without 25 μM DTPA treatment. Genes of interest were normalized to Tbp. Biological replicates, n=3. Values represent mean ± SD. P- values were calculated by two-way ANOVA followed by Tukey’s HSD post-hoc test. Data connected by lines are significantly different at P < 0.05. *Abbreviation: β-Est, β-estradiol; DTPA, diethylenetriamine pentaacetate; siCON, Control siRNA-transfected cells; siZIP10, ZIP10 siRNA-transfected cells. 101 Figure 3-18. Venn diagrams of differentially expressed genes by zinc restriction and ZIP10 depletion during red cell development. Venn diagrams show the numbers of DEGs by DTPA and siZIP10 treatment. DEGs were filtered by the criteria of |Fold change| ≥ 2.0 and adjusted false discovery rate (FDR) P-value < 0.05. **Abbreviation: DEGs, differentially expressed genes; siZIP10, ZIP10 siRNA-transfected cells.DTPA, diethylenetriamine pentaacetate. 102 Table 3-3. Affected canonical pathways by ZIP10 depletion and zinc restriction during red cell development. Each treatment was compared to the siCON under adequate zinc condition via Ingenuity Pathway Analysis (IPA). The canonical pathways were selected by the criteria of P-value < 0.01 and Z-score >1 (Filter: All species, All cell type) in at least one group. Z-scores of each canonical pathway are shown in the table. siCON and siZIP10 indicate control- and ZIP10- siRNA transfection, respectively. ZnA and ZnD indicate adequate zinc status and zinc deficiency, respectively. Canonical Pathways (siCON −DTPA vs.) Z-score siCON + ZnD siZIP10 + ZnA siZIP10 + ZnD Kinetochore Metaphase Signaling Pathway -2.138 -2.6 -3.43 Cell Cycle Control of Chromosomal Replication N/A -2.646 -4.583 Cyclins and Cell Cycle Regulation -2 -2.236 -1.941 Cardiac Hypertrophy Signaling (Enhanced) 1.342 N/A 4.636 Breast Cancer Regulation by Stathmin1 1.265 0.707 3.709 Senescence Pathway 1.414 1.134 3.053 Cardiac Hypertrophy Signaling N/A N/A 5 Cell Cycle: G2/M DNA Damage Checkpoint Regulation 2.236 1.667 0.832 Hepatic Fibrosis Signaling Pathway 1.342 N/A 3.307 Mitotic Roles of Polo-Like Kinase -1.134 -1.667 -1.5 Coronavirus Pathogenesis Pathway 2 N/A 2.294 Role of NFAT in Cardiac Hypertrophy N/A N/A 4.146 Role of CHK Proteins in Cell Cycle Checkpoint Control 2.236 1.134 0.632 Gα12/13 Signaling N/A N/A 4 Role of BRCA1 in DNA Damage Response -0.447 -1.667 -1.807 GP6 Signaling Pathway N/A N/A 3.873 Estrogen-mediated S-phase Entry N/A -2 -1.667 Superpathway of Cholesterol Biosynthesis N/A N/A -3.606 Tumor Microenvironment Pathway N/A N/A 3.411 Colorectal Cancer Metastasis Signaling N/A N/A 3.3 Adrenomedullin signaling pathway N/A N/A 3.3 P2Y Purigenic Receptor Signaling Pathway N/A N/A 3.207 Renin-Angiotensin Signaling N/A N/A 3.162 NER (Nucleotide Excision Repair, Enhanced Pathway) N/A N/A -3.153 Calcium Signaling N/A N/A 3.153 Signaling by Rho Family GTPases N/A N/A 3 RANK Signaling in Osteoclasts N/A N/A 3 cAMP-mediated signaling 1 N/A 1.826 B Cell Receptor Signaling N/A N/A 2.711 White Adipose Tissue Browning Pathway N/A N/A 2.673 Osteoarthritis Pathway N/A N/A 2.673 Cardiac β-adrenergic Signaling N/A N/A 2.668 (Table continues on next page) 103 Canonical Pathways (siCON + ZnA vs.) Z-score siCON + ZnD siZIP10 + ZnA siZIP10 + ZnD IL-1 Signaling N/A N/A 2.646 Superpathway of Geranylgeranyldiphosphate Biosynthesis I (via Mevalonate) N/A N/A -2.646 Leukocyte Extravasation Signaling N/A N/A 2.558 Rac Signaling N/A N/A 2.53 ILK Signaling N/A N/A 2.524 GNRH Signaling N/A N/A 2.524 Thrombin Signaling N/A N/A 2.5 PD-1, PD-L1 cancer immunotherapy pathway N/A N/A -2.496 Relaxin Signaling N/A N/A 2.496 Cholesterol Biosynthesis II (via 24,25-dihydrolanosterol) N/A N/A -2.449 IL-9 Signaling N/A N/A 2.449 iNOS Signaling N/A N/A 2.449 Cholesterol Biosynthesis I N/A N/A -2.449 Cholesterol Biosynthesis III (via Desmosterol) N/A N/A -2.449 CD27 Signaling in Lymphocytes N/A N/A 2.449 Mevalonate Pathway I N/A N/A -2.449 FAT10 Cancer Signaling Pathway N/A N/A 2.449 Salvage Pathways of Pyrimidine Ribonucleotides -0.447 -2 0 PI3K Signaling in B Lymphocytes N/A N/A 2.357 MIF Regulation of Innate Immunity N/A N/A 2.333 p38 MAPK Signaling N/A N/A 2.324 Melanocyte Development and Pigmentation Signaling N/A N/A 2.309 STAT3 Pathway N/A N/A 2.309 Sperm Motility N/A N/A 2.309 α-Adrenergic Signaling N/A N/A 2.236 TNFR2 Signaling N/A N/A 2.236 MIF-mediated Glucocorticoid Regulation N/A N/A 2.121 Opioid Signaling Pathway 0 N/A 2.043 B Cell Activating Factor Signaling N/A N/A 2 Regulation of Cellular Mechanics by Calpain Protease N/A N/A 2 Pyridoxal 5'-phosphate Salvage Pathway N/A -2 0 4-1BB Signaling in T Lymphocytes N/A N/A 2 ATM Signaling 1.342 -0.378 -0.258 Estrogen-Dependent Breast Cancer Signaling N/A N/A 1.89 HER-2 Signaling in Breast Cancer N/A N/A 1.877 Erythropoietin Signaling Pathway 1.342 N/A 0.426 Wnt/Ca+ pathway N/A N/A 1.667 Pancreatic Adenocarcinoma Signaling N/A N/A 1.633 IL-23 Signaling Pathway N/A N/A 1.633 Protein Kinase A Signaling 0.447 N/A 1.093 Dilated Cardiomyopathy Signaling Pathway N/A N/A 1.508 (Table continues on next page) 104 Canonical Pathways (siCON + ZnA vs.) Z-score siCON + ZnD siZIP10 + ZnA siZIP10 + ZnD Ferroptosis Signaling Pathway N/A N/A 1.5 14-3-3-mediated Signaling N/A N/A 1.414 Ceramide Signaling N/A N/A 1.414 Type II Diabetes Mellitus Signaling N/A N/A 1.414 Pyrimidine Deoxyribonucleotides De Novo Biosynthesis I N/A N/A -1.342 Notch Signaling N/A N/A 1.342 Cell Cycle Regulation by BTG Family Proteins N/A N/A -1.342 tRNA Splicing N/A N/A 1.265 Th2 Pathway N/A N/A 1.213 Small Cell Lung Cancer Signaling N/A N/A 1 PI3K/AKT Signaling N/A N/A 1 105 Figure 3-19. ZIP10 depletion reduces cell viability under mild zinc restriction during the terminal erythroid differentiation. Cell viability was measured at 24 and 48 hours of the differentiation in siCON- or siZIP10- transfected G1E-ER4 cells under mild zinc deficiency (25 µM DTPA) using automatic cell counter R1 (Olympus). Biological replicates, n=4 for the 24-hour experiment; Independent experiments, n=3 for the 48-hour experiment. Values represent mean ± SD. P-values were calculated by two-way ANOVA followed by Tukey’s HSD post-hoc test. Data connected by lines are significantly different at P < 0.05. **Abbreviations: β-Est, β-estradiol; DTPA; siCON, Control siRNA-transfected cells; siZIP10, ZIP10 siRNA-transfected cells.diethylenetriamine pentaacetate. 106 Table 3-4. Differential expression of ferroptosis-related genes by ZIP10 depletion and zinc restriction in developing G1E-ER4 cells. Differentially expressed ferroptosis-related genes in the dataset from the Ingenuity Pathway Analysis (IPA) were identified by the criteria of |Fold change| > 2.0 and FDR P-value < 0.01 (Filtered; All species, All cell type). The Fold changes of differentiating G1E-ER4 cells with zinc restriction and ZIP10 depletion compared to control are presented in the table. **Abbreviations: Abca1, ATP-binding cassette subfamily A member 1; Sat1, spermidine/spermine N1- acetyltransferase 1; Fdft1, farnesyltransferase 1; Yap1, yes-related protein 1; Hmgcr, 3-Hydroxy-3- Methylglutaryl-CoA Reductase; H2az1, H2A.Z Variant Histone 1; Slc1a5, solute carrier family 1 member 5; Rbl1, RB (retinoblastoma) transcriptional corepressor like 1; Fancd2, Fanconi anemia complementation group D2; N/A, not available. Gene Name Fold change by ZIP10 depletion and zinc restriction Reported role and involved mechanisms Association with Ferroptosis Reference Abca1 +4.86 Mediating cholesterol efflux to ER; the mevalonate pathway + (150–152) Sat1 +2.18 Inducing lipid peroxidation, a target of p53; polyamine metabolism + (153) Fdft1 −2.70 Squalene synthase (SQS); the mevalonate pathway − (154,155) Yap1 -2.11 Hippo pathway effector + (156) Hmgcr -2.07 HMG-CoA reductase; the mevalonate pathway − (152,157) H2az1 -2.05 DNA repair and genome stability, inducing cystine- glutamate antiporter, Slc7a11, by ubiquitination N/A N/A Slc1a5 -2.68 L-glutamine uptake; Glutaminolysis pathway − (158) Rbl1 -3.26 Cell cycle suppression N/A N/A Fancd2 -17.52 DNA repair − (159) 107 Discussion Cellular zinc homeostasis is tightly managed through tissue-specific zinc transporters. On the plasma membrane of erythrocytes, a zinc exporter, ZnT1, and zinc importers, ZIP8 and ZIP10, were detected in mice (85), implying their potential role in erythroid zinc homeostasis. Recently, the molecular and cellular roles of erythroid ZIP8 and ZnT1 in erythroid zinc management were described using G1E-ER-GATA1 cells and primary mouse and human erythroid precursor cells (8). Yet, the physiological importance of erythroid ZIP10 in the erythron has not been well-characterized. In the present study, we demonstrated the regulation of erythroid ZIP10 by cellular zinc status and the requirement of erythroid ZIP10 for normal erythroid maturation when the extracellular zinc availability becomes suboptimal. ZIP8 has been demonstrated essential for normal hematopoiesis using live animal models of ZIP8 deficiency (86). A mutation leading to a loss in Zip8 activity has been demonstrated to cause dysregulated hematopoiesis manifested as severe anemia in mice (86). Moreover, Zip8 is a regulatory target of GATA1 activity which is induced early during the differentiation of erythroid progenitors (8). However, the expression of ZIP8 appears to be spared by either the zinc or heme status of erythroid cells (8). In addition, a ZIP8 protein abundance from the plasma membrane fractions of erythrocytes was not affected by a zinc-deficient diet in mice (85). These suggest that the transporter is unlikely to be involved in any compensatory mechanisms against cellular zinc or heme deficiency. ZnT1 expression is upregulated by GATA1 upon the development of erythroid progenitors, and it is amplified as cells accumulate heme by terminal erythroid 108 differentiation (8). Zinc status also can regulate ZnT1 expression via the primary zinc- sensing mechanism for the maintenance of cellular zinc homeostasis, MTF-1-MRE- mediated transcriptional regulation in nonerythroid cells (52,160). In mice, a zinc- restricted diet has been shown to reduce the levels of ZnT1 on a plasma membrane of mature RBC (85). In our studies, we found that ZnT1 transcript abundance responded to acute zinc depletion produced by a membrane-permeable zinc chelator, TPEN. However, ZnT1 expression in differentiating erythroid cells was not repressed by the treatment of a cell membrane-impermeable zinc chelator, DTPA. This difference may be due to the difference in the degrees of impact of each zinc chelator on the known regulators for ZnT1, which include GATA1 activity (8), heme status (8), and the availability of zinc (85). Unlike erythroid ZIP8 and ZnT1, ZIP10 transcript abundance rapidly drops in G1E-ER4 cells as the cells initiate terminal erythroid differentiation when the cellular zinc supply is normal. ZIP10 regulation has been identified to be mediated by zinc status via MTF-1 mediated transcriptional regulation in mice and zebrafish (52,161). The MTF- 1-zinc complex binds to the MRE of the ZIP10 gene, which is downstream to the transcription start site, to hinder the movement of polymerase II. This results in the transcriptional repression of Zip10 in response to a rise in cytosolic zinc (52). Conversely, cellular zinc depletion alleviates the physical interference of MTF-1 complex by inhibiting the nuclear translocation of the transcription factor, and thus prevents its binding to the MRE of Zip10 (52). This mode of ZIP10 regulation may function as the regulatory mechanism by which ZIP10 responds in developing erythroid progenitors, and may account for the increased ZIP10 protein on the plasma membrane of mature RBCs in 109 mice fed a low zinc diet (85). In the present study, ZIP10 transporter expression was not affected by cellular heme deficiency produced by iron or PPIX restriction. This is in agreement with a previous observation showing no effects of ALAS2 deficiency on ZIP10 expression (8). Our in vitro loss-of-function study indicates that ZIP10 becomes essential for normal heme biosynthesis only when the cellular zinc supply is limited. Thus, ZIP10 might be dispensable for adequate erythroid development when systemic zinc levels are sufficient, but becomes essential for the maintenance of cellular zinc homeostasis by upregulating its expression, and thereby facilitating cellular zinc uptake, when erythroid cells encounter a zinc-restricted environment. There is no reliable biomarker for the prediction of zinc deficiency in humans (162), which limits the identification of patients who may benefit from zinc treatments for therapy or prevention of particular zinc-related disorders. Plasma/serum zinc, urinary zinc, zinc concentration in erythrocytes, platelets or lymphocytes, and zinc metalloenzyme activity such as ALAD and alkaline phosphatase activity in erythrocytes have been proposed as potential biomarkers of zinc status (137,163–166). Plasma/serum zinc is commonly assessed as a biomarker for zinc status in dietary zinc interventions where dietary zinc intake and other potential factors affecting blood zinc are strictly controlled (167,168). However, plasma/serum zinc levels are not suitable for the diagnosis of inadequate zinc status as it is sensitive enough for identifying marginal deficiency status (169), and plasma/serum zinc levels can be affected by non-zinc factors, infection, inflammation, stress, and even the time of blood collection (170). RBCs have been used as specimens for diagnostic purposes due to the relatively non-invasive nature of sample collection. For instance, in clinical settings, RBC magnesium has been 110 assessed to diagnose magnesium status (171). For zinc assessment, molecule indices, including erythrocyte metallothionein (172) and erythrocyte membrane zinc (173), have been proposed as biomarkers for human zinc deficiency, but the reliability and effectiveness of those biomarkers remain unclear (166). The current study suggests that erythroid ZIP10 responds to zinc restriction in the erythron, including developing erythroid cells from our findings and mature RBCs (85). Moreover, the response of erythroid ZIP10 appears relatively specific to zinc restriction and resistant to changes in iron availability or cellular heme contents. Further studies are needed to investigate whether other physiological cues, such as inflammation, affect erythroid ZIP10. With further validation of the specificity of erythroid ZIP10 to zinc status, erythroid ZIP10 could serve as a diagnostic biomarker, in addition to the ZPP/Heme ratio that we have proposed in the previous chapter. Zinc import by ZIP10 has been reported in different organs. In brain cells and hepatocytes, elevated ZIP10 was reported in response to zinc depletion (52). In nonerythroid cells, the essential role of ZIP10 in zinc homeostasis has been described employing conditional Zip10 knock-out mice models. A deletion of ZIP10 in early B-cell development, including pro-B and pre-B cells, and mature B-cells reduced intracellular zinc levels and their survival and immune responses (17,174). Conditional ablation of Zip10 in macrophages of mice decreased the number of macrophages when animals were challenged by LPS-induced inflammatory responses (127). This was accompanied by a lower accumulation of intracellular zinc (127), indicating how ZIP10 and zinc homeostasis are important during development when the demand for intracellular zinc increases. Additionally, conditional deletion of Zip10 impaired in the epidermal 111 progenitors of mice impaired the progression of epidermal development, and thus caused epidermal hypoplasia (175). This ablation of Zip10 caused a failure in the zinc influx to promote p63 transactivation in epidermal progenitor cells, which mediates epidermal proliferation and differentiation. The present study reveals the essential role of erythroid ZIP10 via in vitro models of developing erythroid progenitors, which do not fully differentiate into mature red cells (116). Previous dietary zinc studies with rodents suggest a role of zinc in erythrocyte membrane integrity and lifespan (1,35). Development and phenotypic characterization of an erythroid-specific conditional Zip10 knockout model will further validate our current mechanistic studies on the roles of ZIP10 during zinc deficiency and will allow for in-depth investigation of the importance of the zinc transporter in the functioning or survival of mature erythrocytes in circulation. Differentiating erythroid progenitor cells manage a large amount of iron due to heme production, and thus mishandling iron could induce oxidative stress and redox activity (146). Recently, this potential cytotoxic effect of iron has been identified as the cause of an iron-dependent nonapoptotic regulated cell death, ferroptosis (176). The current study revealed a reduction in cell viability produced by the losses of both cellular zinc and ZIP10 and identified the ferroptosis signaling pathway as an affected canonical pathway modulated by cellular zinc and ZIP10 status. Of particular relevance is the mevalonate pathway the which is also identified via IPA canonical pathway analysis (Table 3-3). The detected DEGs in the ferroptosis signaling pathway include genes also involved in the mevalonate pathway that mediates cholesterol synthesis through a cascade of several enzymatic reactions. The mevalonate pathway is essential for the maturation of the anti-oxidative enzyme, glutathione peroxidase 4 (GPX4), which plays a 112 role in the prevention of ferroptosis (177). Increased ABCA1 can mediate cholesterol efflux to the endoplasmic reticulum and thus can inactivate SREBP2, a transcription factor for the suppression of gene expressions involved in the mevalonate pathway(150). Inactivated SREBP2 suppresses the accumulation of squalene, an intermediate of cholesterol synthesis. A reduction in HMGCR (178) and FDFT1 (squalene synthase, SQS) also prevents squalene accumulation (154,155). This failure of squalene accumulation potentially results in the induction of ferroptosis by oxidative damage on the membrane (154). Moreover, FANCD2, a nuclear protein involved in the repair of DNA strand damage, plays a role in the defense against ferroptosis in the bone marrow stromal cells (159), which was identified as the most significantly reduced DEG among the ferroptosis-associated genes in the current study. In accordance with these findings, the transcriptome profile of zinc-deficient erythroid progenitors from the previous chapter identifies the DEGs, including increased Abca1, and decreased Hmgcr, Fdft1, and Fancd2, further supporting the role of zinc in the regulation of the antioxidant system and ferroptosis. Collectively, altered zinc homeostasis in the erythroid cells may influence the cellular capacity for defense against or activation of ferroptosis, which warrants further investigation. To our knowledge, this is the first study presenting the importance of ZIP10 in cellular zinc homeostasis during red blood cell development. We identify ZIP10 as a zinc transporter highly responsive to zinc status in developing erythroid progenitors, and such response persists throughout development. Notably, ZIP10 deficiency alone did not produce any phenotypes unless cells were challenged by zinc restriction, suggesting that ZIP10 could be conditionally essential for normal red cell development. In addition to the 113 previous study identifying the potential role of zinc in normal erythroid maturation and heme biosynthesis, our findings suggest ZIP10 mutations along with inadequate zinc status and might serve as potential risk factors for anemia or other disorder of impaired erythroid heme metabolism. 114 Chapter 4 . The identification of genetic variants of ZIP10 among the African population 115 Introduction Zinc deficiency is a nutrition disorder with an estimated global prevalence of approximately 31 %, and anemia is another health concern worldwide, affecting approximately 23% of the population (20,66). Previous studies have demonstrated that zinc deficiency and anemia are associated, and Africa has the highest prevalence of both public health problems (1,34). Considering that we have demonstrated that ZIP10 depletion can impair hemoglobinization when zinc supply is limited, we questioned whether ZIP10 mutations are present in human populations, particularly those that might experience higher risks of zinc deficiency and anemia. By screening through the gnomAD browser on the reference genome GRCh38p.13, three notable missense variants of ZIP10 (SLC39A10) specific to the African population have been identified by our laboratory; rs13419724, rs114992984, and rs76182486. If the ZIP10 variants are hypomorphic, individuals with the mutation are expected to contain less zinc in their erythrocytes, possibly leading to erythroid-related disorders, such as anemia. Thus, the aim of this study was to experimentally validate the presence of African-centric ZIP10 variants identified via database screening and to predict the potential functional implications of the ZIP10 variants. The findings may suggest the potential association between ZIP10 variants among the African population and the high prevalence of endemic erythrocyte-related diseases. 116 Materials and Methods SNP screening through the database The gnomAD database (https://gnomad.broadinstitute.org) was utilized to identify ZIP10 SNPs (179). This includes whole-exome sequencing data of 123,136 exomes and 15,496 genomes of unrelated individuals sequenced from several large-scale projects. Additional identification of ZIP10 SNPs was obtained from the 1000 Genomes Project phase 3, representing 2504 individuals on GRCh.37 (http://grch37.ensembl.org/index.html). Combined annotation-dependent depletion (CADD), a tool for measuring deleteriousness of gene variants that have potential changes on its functions, was identified through the CADD - Combined Annotation Dependent Depletion website (https://cadd.gs.washington.edu) (180). The population- based distributions of each SNP in the 1000 Genomes Project phase 3 were generated through the Geography of Genetic Variants Browser (https://popgen.uchicago.edu/ggv/) (181). Predicted topology of ZIP10 and the location of its mutations The predicted transmembrane protein topology of ZIP10 was generated via the MEMSAT/SVM (http://bioinf.cs.ucl.ac.uk/psipred/) (182). For evolutionary amino acid sequence alignment, ZIP10 amino acid sequences of humans, chimpanzees, mice, and rats were aligned using Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/). Whole blood and DNA sample preparation of the African population Whole blood samples of Ugandan children (n=100; aged 59 months to 6 years) were obtained from Dr. Sarah Cusick at the University of Minnesota in the department of 117 pediatrics (183). DNA was extracted from the whole blood samples using the Quick- DNA Miniprep kit (Zymo Research) according to the manufacturer’s protocol. Another large set of DNA samples from the African population was provided by Dr. Chandy C. John from the Indiana University School of Medicine (n=655) from his previous study. Genotyping qPCR-based SNP genotyping was performed using rhAmp SNP Assays T_rs13419724, A_rs114992984, and T_rs76182486 according to the manufacturer’s instruction. Approximately 10 ng of each DNA sample was used for the assay. Allelic discrimination plots of genotype assignment were determined by plotting the endpoint relative fluorescent units (RFU) of FAM for the reference allele on the x-axis and VIC® dye for the alternate allele on the y-axis. Genotype-based whole blood assessment Hemoglobin concentrations from the Ugandan children were found in the published study (183). Zinc protoporphyrin from the whole blood sample was measured as previously described in Chapter 2. The results were grouped by the ZIP10 SNP genotypes. Student’s t-test was conducted to define significant differences. A P-value < 0.05 was considered statistically significant. Results The three notable ZIP10 SNPs specific to the African population are identified through public databases. By screening through the SLC39A10/ZIP10 variants deposited in the gnomAD (version 2.1.1.) and 1000 Genome Project Phase 3 database, we identified three notable 118 missense variants of ZIP10 specific to the African population; rs13419724, rs114992984, and rs76182486. Their amino acid and allele changes, allele frequencies, and CADD scores by population according to the gnomAD (version 2.1.1.) and 1000 Genome Project Phase 3 database are summarized in Table 4-2 and Table 4-1, respectively. The population-based distributions of allele frequencies from the 1000 Genomes project were visualized on the world maps. This visually confirmed that these SNPs are highly present in Africans and African ancestry compared with non-Africans (Figure 4-1). The SNP A>T allele substitution at rs13419724 results in the replacement from threonine to serine, both containing uncharged polar side chain, at the amino acid position 87 (Thr87Ser), and the SNP C>A allele substitution at rs114992984 changes to the amino acid from the electrically charged side chain histidine to the uncharged polar glutamine at the position 194 (His194Gln). The SNP C>T allele variant at rs76182486 results in the substitution of the electrically charged histidine to the hydrophobic tyrosine at the amino acid position 609 (His609Tyr). These missense mutations may lead to changes in protein structure, which could result in phenotypic changes, and thus we applied combined annotation- dependent depletion (CADD) score to computationally predict the deleteriousness of gene variants that may have significant effects on its functions (180). For instance, a CADD score of 10 and 20 indicates that the variant is in the top 10% and 1% of deleterious variants among human genomes, respectively. The CADD scores of each ZIP10 SNP were about 10 to 20. Notably, the His609Tyr substitution (rs76182486) has the highest CADD score of 22 among the identified three ZIP10 SNPs. 119 The topology of human ZIP10 and the location of its mutations in the evolutionary alignment of the gene Human ZIP10 protein encoded by SLC39A10 is in the LIV-1 subfamily that is a highly conserved group of zinc transporter containing eight transmembrane domains (184–187). All members of the SLC39A LIV-1 subfamily share several common sequences. The consensus amino acid sequences placed in the domains between TM IV and V, a histidine-rich motif on the long cytoplasmic loop between TM III and IV, and a metalloprotease-like motif in TM V are the motifs in common in the LIV-1 family (10,184). Importantly, the histidine-rich motif on the intracellular loop between TM III and IV is predicted to be important for zinc transport (184,188). ZIP10 protein contains 49 histidines in the extracellular N-terminus, 9 histidines on the extracellular loop between TM II and III, and 20 histidines on the cytoplasmic loop between TM III and IV (184,185). The predicted topology of human ZIP10 using the MEMSAT/SVM and the location of the three SNPs are presented in Figure 4-2. The predicted model of human ZIP10 includes six transmembrane (TM) helices, and three cytoplasmic domains, and four extracellular domains. However, since the present study predicted six transmembrane domains via MEMSAT/SVM, which does not follow the classic structure of eight TM in the LIV-1 family, the positions of predicted TM VI and V were illustrated based on the published structural features of ZIP transporters (189,10,186). The locations of the Thr87Ser (rs13419724) and His194Gln (rs114992984) in human ZIP10 are in the 120 extracellular region near NH2-terminal, whereas His609Tyr (rs76182486) variant is located in the cytoplasmic region between the TM domains III and IV. Most disease-causing mutations have been associated with missense mutations resulting in changes of amino acids at evolutionarily conserved regions (190,191). Thus, we compared the ZIP10 amino acid sequence of mammals, chimpanzees, mice, and rats, with the human ZIP10 protein (Figure 4-3). The corresponding orthologous positions for the three SNPs are indicated on the alignment. Notably, we observed that the Thr87Ser and His609Tyr variants result in changes of amino acids which have been evolutionarily highly conserved, whereas the His194Gln variant appears on the relatively less conserved amino acid. Our analyses on the sequences of the African-centric ZIP10 variants suggest that the Thr87Ser and His609Tyr variants occurred in a highly conserved region of the protein and thus may have functional implications. The ZIP10 SNPs are present in the African population The three missense ZIP10 SNPs centric to Africa were initially identified by screening a database of whole genome and exome sequences. Among those, we experimentally confirmed the presence of rs13419724, which has the highest minor allele frequency, with the Ugandan children samples (n=100). The genotyping success rate was 98 %, including one no-called sample and one compromised whole blood sample (Figure 4-4). As shown in Table 4-3, the allele frequency of the alternative allele (T) was 6.63%, which was shown as 8.62% and 7.96% in the African population from 1000 Genome Project Phase 3 and gnomAD, respectively. This data experimentally validated that the Thr87Ser variant encoded by rs13419724 is present in the African population. 121 Additional genotyping with a larger number of African DNA samples was conducted to identify the ZIP10 SNPs with lower minor allele frequencies. The genotyping success rate was 100 % for all three ZIP10 SNPs (Figure 4-5, Figure 4-6, Figure 4-7). As shown in Table 4-4, we experimentally validated the minor allele frequencies of all ZIP10 SNPs above 1%. Collectively, in accordance with the previous data from the 1000 Genome and gnomAD database, this data showed the presence of all three ZIP10 SNPs in the African population and identified that the Thr87Ser variant is the most prevalent allele among the three missense ZIP10 SNPs. ZIP10 SNP specific to the African population is associated with hematological indices in the Ugandan children The data from Cusick et al.’s study (183) and whole blood samples provided from the lab allowed us to further investigate ZIP10 genotyped-based hematological implications in the African population. The heme data (183) and newly measured zinc protoporphyrin in the whole blood samples from the Ugandan children were grouped based on ZIP10 SNP (rs13419724, Thr87Ser, A>T) genotype (Figure 4-8 and Table 4-5). Hemoglobin concentrations were significantly higher in ZIP10 SNP (Thr87Ser) children when compared to levels of patients with the common allele. We hypothesize that the ZIP10 SNP could be hypomorphic, leading to lower erythroid zinc import, and thus ZnPP concentrations were measured using frozen whole blood samples from each patient. The genotype effect on blood ZnPP concentrations was not statistically significant. However, a trend toward a lower ZPP/H ratio by the presence of the ZIP10 SNP (Thr87Ser) was observed, which was likely driven by the significantly higher heme content. 122 Table 4-1. Three missense genetic variants specific to the African population in the SLC39A10/ZIP10 gene (gnomAD database). Gene SNP ID Allele frequency African/African American Ashkenazi Jewish Latino/Admixed American Other European (non- Finnish) South Asian East Asian European (Finnish) SLC39A10 rs13419724 0.0796 0.0084 0.0045 0.0044 0.0005 0.0002 0.0000 0.0000 rs114992984 0.0154 0.0006 0.0006 0.0002 0.0001 0.0001 0.0000 0.0000 rs76182486 0.0074 0.0003 0.0002 0.0000 0.0000 0.0000 0.0000 0.0000 The allele frequencies are identified from the gnomAD v2.1.1 through the gnomAD browser. 123 Table 4-2. Three missense genetic variants specific to the African population in the SLC39A10/ZIP10 gene (1000 Genome Project). Positions were presented based on GRCh37.p13. The Allele frequencies are from the 1000 Genomes Project Phase 3 database. *Abbreviation; EUR, European (non-Finnish and Finnish); EAS, East Asian; SAS, South Asian; AFR, African; AMR, American. Gene SNP ID Position Allele Allele frequency Amino acid CADD score EUR EAS SAS AFR AMR SLC39A10 rs13419724 chr2:196545025 A>T 0.0000 0.0000 0.0000 0.0862 0.0058 Thr87Ser 18.50 rs114992984 chr2:196545348 C>A 0.0000 0.0000 0.0000 0.0204 0.0029 His194Gln 10.58 rs76182486 chr2:196581489 C>T 0.0000 0.0000 0.0000 0.0098 0.0000 His609Tyr 22.20 124 Figure 4-1. The population-based distribution map of the three ZIP10 SNPs which are specific to the African population. The minor allele frequencies of ZIP10 (A) rs13419724, (B) rs114992984, and (C) rs76182486 from the 1000 Genomes (hg19) database are shown in each pie. Each pie indicates populations investigated in the database. The blue and green portion of the pie represents a minor allele frequency out of 1 and 0.1, respectively. The abbreviations of the populations containing minor alleles are shown. **Abbreviations: ASW, African Ancestry in SW USA; ACB, African Caribbean in Barbados; CLM, Colombian in Medellín, Colombia; ESN, Esan in Nigeria; GWD, Gambian in Western Division – Mandinka; LWK, Luhya in Webuye, Kenya; MSL, Mende in Sierra Leone; PEL, Peruvian in Lima Peru; PUR, Puerto Rican in Puerto Rico; YRI, Yoruba in Ibadan, Nigeria. 125 Figure 4-2. A predicted topological model of human SCL39A10/ZIP10. The predicted topology of human ZIP10 is present with six transmembrane helices, and three cytoplasmic domains, and four extracellular domains, including NH2- and COOH-terminus. The transmembrane (TM) domains are numbered I–VIII. The gray TM and the dotted lines indicate reported topology, which is not presented in the prediction using the MEMSAT/SVM but identified in previous studies (10). The orthologous positions of the Thr87Ser (rs13419724), His194Gln (rs114992984), and His609Tyr (rs76182486) substitutions in human ZIP10 are indicated in green, yellow, and orange, respectively. The predicted pore-lining region in transmembrane proteins is the position 761-821. 126 Figure 4-3. Evolutionary alignment of human, chimpanzee, mouse, and rat ZIP10 amino acid sequences with the variants of interest. Amino acid sequence conservation of ZIP10 in mammals including humans, chimpanzees, mice, and rats is shown using Clustal Omega. An asterisk (*) indicates residues that are identical to human ZIP10, and a colon (:) and a period (.) present a strong and weak similarity of each aligned amino acid properties, respectively. Prediction of the transmembrane structures was produced using MEMSAT/SVM. Extracellular regions, transmembrane helices, and cytosolic regions are indicated in gray, red, and white, respectively. The positions of the human rs13419724, rs114992984, and rs76182486 ZIP10 SNPs are indicated in the corresponding amino acid positions. The amino acid sequences used to generate the evolutionary alignment include NP_065075.1 (Human; Homo sapiens), XP_016805738.1 (Chimpanzee; Pan troglodytes), NP_766241.2 (Mouse; Mus musculus), and NP_001102266.2 (Rat; Rattus norvegicus). 127 Figure 4-4. An allelic discrimination plot of rhAmp SNP Assays targeting SLC39A10 SNP, rs13419724 among Ugandan children. DNA samples from the Ugandan children (n=100) were genotyped using rhAmp ADME SNP Assay targeting SNP rs13419724 [A/T]. The A/A, AT, and T/T represent the ZIP10 common allele homozygous, heterozygous individuals, ZIP10 SNPs (rs13419724) homozygous individuals, respectively. **Abbreviations: RFU, relative fluorescence units; NTC, negative control. 128 Table 4-3. Genotyping of ZIP10 SNP (rs13419724) in Ugandan children. Gene SNP ID Allele Genotype frequency Allele frequency SLC39A10 rs13419724 A>T A/A A/T T/T A T 0.8776 0.1122 0.0102 0.9337 0.0663 *Total number of the samples (n=98) 129 Figure 4-5. An allelic discrimination plot of rhAmp SNP Assays targeting SLC39A10 SNP, rs13419724 among the African population. DNA samples from the Ugandan children (n=100) were genotyped using rhAmp ADME SNP Assay targeting SNP rs13419724 [A/T]. The A/A, A/T, and T/T represent the ZIP10 common allele homozygous, heterozygous individuals, ZIP10 SNPs (rs13419724) homozygous individuals, respectively. **Abbreviations: RFU, relative fluorescence units; NTC, negative control. 130 Figure 4-6. An allelic discrimination plot of rhAmp SNP Assays targeting SLC39A10 SNP, rs114992984 among the African population. DNA samples from the Ugandan children (n=100) were genotyped using rhAmp ADME SNP Assay targeting SNP rs114992984 [C/A]. The C/C, C/A, and A/A represent the ZIP10 common allele homozygous, heterozygous individuals, ZIP10 SNPs (rs114992984) homozygous individuals, respectively. **Abbreviations: RFU, relative fluorescence units; NTC, negative control. 131 Figure 4-7. An allelic discrimination plot of rhAmp SNP Assays targeting SLC39A10 SNP, rs76182486 among the African population. DNA samples from the Ugandan children (n=100) were genotyped using rhAmp ADME SNP Assay targeting SNP rs76182486 [C/T]. The CC and CT represent the ZIP10 common allele homozygous and heterozygous individuals, respectively. **Abbreviations: RFU, relative fluorescence units; NTC, negative control. 132 Table 4-4. Genotyping of ZIP10 SNPs in the African population. Gene SNP ID Allele Genotype frequency Allele frequency SLC39A10 rs13419724 A>T A/A A/T T/T A T 0.8702 0.1252 0.0046 0.9328 0.0672 rs114992984 C>A C/C C/A A/A C A 0.9603 0.0382 0.0015 0.9794 0.0206 rs76182486 C>T C/C C/T T/T C T 0.9725 0.0275 0.0000 0.9863 0.0137 *The total number of the African DNA sample is 655. 133 Figure 4-8. The implication of ZIP10 SNP (rs13419724) on blood levels of hemoglobin and zinc protoporphyrin in the Ugandan children (A) Hemoglobin levels (n=98), (B) zinc protoporphyrin levels (n=96), and (C) ZnPP/Heme ratio (n=96) were measured in the Ugandan children and grouped by ZIP10 SNP (rs13419724) genotypes. Values represent mean ±SD. *P < 0.05 by Student’s t-test. **Abbreviation: ZnPP and ZPP, zinc protoporphyrin. 134 Table 4-5. Hematological indices of ZIP10 SNP (rs13419724) in Ugandan children **Abbreviation: ZnPP and ZPP, zinc protoporphyrin; H, heme; n, number. Biomarkers All, n A/A A/T, T/T P-value n Mean (SD) n Mean (SD) Hemoglobin (g/dL) 98 86 11.3 (1.4) 12 12.6 (0.9) 0.0003 ZnPP (μg/dL) 96 85 15.5 (8.5) 11 12.7 (6.8) 0.2424 ZPP/H (μmol/mol) 96 85 36.7 (22.3) 11 26.8 (15.3) 0.0740 135 Discussion Zinc deficiency is a public health problem with an estimated global prevalence of 31%, and is widespread in low-income populations (192). Inadequate zinc status can be a result of various reasons, including inadequate dietary zinc intake, disturbed absorption by genetic abnormalities, phytate in food sources, and infection (40). Recent studies have suggested that zinc is involved in the survival and propagation of infectious microorganisms (193). Nutritional immunity is a classic innate immune response of the host organism against infections of parasites via the restriction of essential minerals, e.g., iron and zinc, to the pathogens (194). Hypozincemia is an acute phase immune response produced by infection or inflammatory cues (89,96). This is produced primarily via the IL-6-induced activation of hepatic ZIP14, which sequesters blood zinc into tissues such as the liver (89). The decline in blood zinc concentration can prevent further disease progress by pathogenic microorganisms, which require zinc for survival or proliferation. Additionally, deprivation of intracellular labile zinc can serve as an antimicrobial strategy (195). Macrophages phagocytose the pathogens, such as Histoplasma capsulatum, and then can kill them by zinc starvation (195). This is mediated by the increase in MT expression to bind to bioavailable zinc and by upregulation of ZnT4 and ZnT7 to sequester zinc into the intracellular organelle, Golgi apparatus. The sequestration of cytosolic labile zinc limits the growth of the intracellular microbiomes. Thus, these suggest systemic or cellular zinc restriction can function as a strategy of nutritional immunity. All ZIP10 variants identified polymorphic by the gnomAD database (with allele frequencies above 1%) were limited to the African population. Notably, the African 136 populations experience high rates of infectious diseases and carry gene variants which are protective against infection by promoting nutritional immunity mediated by metal redistribution (81,105,106). ZIP4 is the zinc transporter of the apical membrane of brush border cells initiating the absorption of dietary zinc (59). ZIP4 mutations identified to be specific to the African populations have been characterized as hypomorphic, and suggested as a mechanism for protection against infections (106). More recently, a variant of an iron exporter gene FPN1 has been associated with a protective impact against malarial infection, such as less severe symptoms (82). The proposed mechanism is the resistance of hepcidin-induced FPN degradation by the variant, resulting in a less iron supply to the malaria parasites within erythrocytes (81). Erythrocyte zinc experiences an acute and massive expansion when infected by Plasmodium falciparum (107). This expansion in the erythrocytic zinc pool was demonstrated as essential for the growth of the intraerythrocytic parasites during the proliferative stage of development (107). Resident and parasite-synthesized plasma zinc transporters have been postulated as putative routes for the accelerated zinc import in malaria-infected erythrocytes (108). ZIP10 may be of particular relevance as a host cell transporter to this hypothetic model of zinc uptake machinery in the infected erythrocyte (85). Thus, future studies functionally characterizing the human ZIP10 variants and addressing how ZIP10 activity influences in vivo growth of malarial parasites are needed. Recent studies have been proposed zinc transporters as therapeutic targets for diseases. In the T-cell acute lymphoblastic leukemia cell line, ZIP7 has been identified as a target of a compound, NVS-ZP7-4 (196). Inhibition of ZIP7 on the endoplasmic reticulum (ER) membrane by NVS-ZP7-5 elevates zinc levels in ER and thereby induces 137 ER stress and apoptosis via inhibition of the Notch signaling, which is an oncogenic pathway associated with the pathogenesis of T-cell acute lymphoblastic leukemia. ZnT8 also has been discovered to be targeted and inhibited by certain molecules. Haploinsufficiency of human ZnT8 has been associated with type2 diabetes, implying the protective effects of loss of function in ZnT8 against diabetes (19,197). Anti-ZnT8 monoclonal antibodies inhibit ZnT8 transport activity by stabilizing its structural folding, resulting in a partial loss of its function (198). This indicates that the decrease in ZnT8 transport activity by inhibitors may serve as an antidiabetogenic strategy. Additionally, ZIP8 has been identified as a target of tetrahydrocarbazole (S)-3 by screening a database of small organic molecules and confirmed by in vitro studies (199). These studies suggest that zinc transporters can serve as druggable targets for disease treatment, and thus imply that ZIP10 may also serve as a therapeutic target for human diseases. Dysregulation of systemic or cellular zinc homeostasis by genetic variants of zinc transporters has been associated with human diseases. Acrodermatitis enteropathica (AE) is a genetic disorder attributed to a loss-of-function mutation of ZIP4 and is characterized by severe zinc deficiency (59). ZIP4 protein in the small intestine primarily mediates zinc absorption in response to dietary zinc levels (59). The mutations of ZIP4 negatively influence zinc uptake by impairing their localization to the plasma membrane and transport activities (106,200). Notably, this genetically produced zinc deficiency condition can be efficiently corrected by the administration of high doses of dietary zinc intake (98). Similar to this, the ZIP10 mutation may influence an individual’s capability for zinc acquisition and handling, and thus optimal dietary intake. The lack of a specific and reliable biomarker of zinc status has made the identification of patients who may 138 benefit from dietary zinc interventions difficult. ZIP10 genotype could potentially become an index for precision nutrition which aims to deliver individual-centered nutritional recommendations based on background information, such as the genetic variants. In Africa, which has a higher risk of malaria, the genetic variants of ZIP10 SNPs might confer nutritional immunity by influencing zinc homeostasis. However, immigrants, who have carried these ZIP10 SNPs from their ancestral origin, to new locations with a low risk of malaria infection may not benefit from nutritional immunity, and may even suffer from zinc deficiency, if they are provided current guidelines on daily zinc intake for the general population. In this regard, precision nutrition may improve the zinc status through tailored recommendations of daily zinc intake based on the genetic variants instead of such “one-size-fits-all” guidelines (201). Our previous studies have demonstrated that ZIP10 may be conditionally essential for erythroid zinc homeostasis by mediating zinc uptake when cells encounter a zinc- restricted environment. The present whole blood analysis from the Ugandan children showed the higher hemoglobin concentration by the Thr87Ser substitution of ZIP10. This observation could be speculated with several explanations. The advantages of ZIP10 on hemoglobinization were only observed when cells were challenged by zinc restriction. Under normal zinc conditions, developing erythroid cells repressed ZIP10 expression, indicating their preference for lower ZIP10 activity. This can be supported by the potential interference of iron incorporation into protoporphyrin by excessive zinc towards the later stage of development (115). Additionally, the mutation of ZIP10 may affect different cell types, resulting in high hemoglobin as a secondary result. For instance, conditional ablation of Zip10 in B-cells decreases cellular zinc homeostasis and thus 139 impairs cell survival and immune responses (17,174). Loss of Zip10 in macrophages under LPS-induced inflammatory stimulation also reduces the number of macrophages in mice (127), which could impair red blood cell recycling via erythrophagocytosis (202). This may cause high hemoglobin in the whole blood. Lastly, the mutations may provide gain-of-function. Zinc transporter mutations can result in metal specificity (203) or altered cellular zinc pool by mislocalization (204). Collectively, the hematological implications of the African-specific ZIP10 SNPs warrant further investigation. Mutations in zinc transporters can affect their localization and zinc transport activity. A histidine-rich motif on the large cytoplasmic loop between TM III and IV is a shared sequence in the SLC39A LIV-1 subfamily (10,184). The histidine-rich motif has been shown to serve as a metal-binding site in an iron transporter in yeast (205) and zinc transporter ZnT4 in rats (206). Mutations on the histidine-rich motif located on the loop between TM III and IV of zinc transporter have been demonstrated to cause a decrease in the zinc uptake rate or mislocalization of the transporter in yeast (207) and human (208) ZIP1. Despite the discrepancy between the predicted human ZIP10 topology in the present study and published reports (189,10,186), we identify that the His609Tyr (rs76182486) substitution is located on the cytoplasmic loop between TM III and IV, which is postulated to be important in zinc transport. Additionally, we found that the His609Tyr substitution, which is mutated from histidine (electrically charged) to tyrosine (hydrophobic), occurs on the highly conserved residue in mammals (Figure 4-3). Collectively, these findings imply that the His609Tyr variant may be functionally hypomorphic and thus impair the import of exogenous zinc into cells. This is in agreement with its high CADD score above 20, meaning the top 1% of deleterious 140 variants in the human genome. Based on these predictions, further investigations are needed to investigate the functional implications of the ZIP10 SNPs, particularly His609Tyr substitution, on its expression, localization, and activity via in vitro study. Collectively, the current study identifies three polymorphic ZIP10 SNPs in the African population, that may experience higher risks of zinc deficiency and anemia. We experimentally validated the presence of these ZIP10 variants in the African population and found a ZIP10 genotyped-based hematological implication in the African population. Our findings suggest the potential impacts of the African-centric ZIP10 SNPs on nutritional immunity and precision nutrition, which warrants further investigation. 141 Chapter 5 . Summary 142 Zinc is a nutrient important for diverse biological processes in the body. Zinc deficiency is a highly prevalent nutrition disorder with an estimated global prevalence of 31 %. As a type 2 nutrient, zinc deficiency manifests as various symptoms. The association between zinc deficiency and anemia in humans has been recognized in several studies since the first study documenting zinc deficiency in humans reported a symptom similar to iron deficiency anemia. Yet, how erythroid zinc status influences red cell development and hemoglobin production, which are critical processes of the etiology of anemia, remains poorly understood. The current studies aim to elucidate the roles of zinc and zinc importer ZIP10 during red cell development and determine the presence of ZIP10 SNPs in the populations that might experience higher risks of zinc deficiency and erythroid-related diseases. The first study mechanistically characterized the requirement and role of zinc in establishing the cellular machinery for heme biosynthesis during terminal erythroid differentiation. G1E-ER4 cells, a mouse erythroid progenitor cell line, were employed as a study model for terminal erythroid differentiation. We determined a significant accumulation of total cellular zinc contents by 1.7-fold in G1E-ER4 cells at 48-hour of differentiation. Restriction of extracellular zinc supply using a membrane-impermeable chelator, DTPA, impaired hemoglobinization prior to any losses in the total number of cells and cell viability, and the decreases in heme contents by DTPA were fully restored by replenishing equimolar zinc. This was accompanied by no changes in total cellular iron contents. Among heme biosynthetic genes, significant decreases in ALAD expressions were observed with normal Alas2 mRNA abundance by zinc restriction in differentiating G1E-ER4 cells. To investigate the production of heme precursor 143 metabolite, protoporphyrin (PPIX), MEL-DS19 cells, a mouse erythroleukemia cell line, were employed. A decrease in PPIX was observed in MEL cells experiencing heme deficiency by zinc restriction. In brief, zinc restriction impairs hemoglobinization of erythroid progenitors, not merely due to impaired induction of differentiation but rather caused by impairments in the metabolic process, heme production. Cellular zinc homeostasis is tightly managed via differential expressions of the cell-type-specific zinc transporters. Yet, the precise roles and requirements of erythroid zinc transporters on zinc homeostasis in terminal erythroid differentiation remain uncertain. The second study aimed to identify a zinc transporter important in maintaining erythroid zinc homeostasis during red cell development and zinc deficiency. Among all 24 zinc transporters, the Zip10 gene was identified as the most responsive gene to zinc restriction in developing G1E-ER4 cells. Zip10 levels were upregulated by zinc restriction but not mediated by heme. Using a siRNA-mediated gene silencing approach, we demonstrated that ZIP10 depletion in developing G1E-ER4 cells had no effect when zinc availability was normal, but exacerbated heme deficiency by zinc restriction. Additionally, acute zinc depletion using a membrane-permeable chelator TPEN and extensive extracellular zinc restriction using a membrane-impermeable chelator DTPA caused losses in the total number of cells and cell viability of G1E-ER4 cells undergoing differentiation. Bioinformatic analysis of the transcriptome profiles in zinc- and ZIP10- deficient G1E-ER4 cells identified the enrichment of gene responses associated with ferroptosis, which may influence the cellular capacity for defense against or activation of ferroptosis. Overall, these findings demonstrate that ZIP10 might be dispensable for normal erythroid development when zinc availability is adequate, but becomes essential 144 for the maintenance of erythroid zinc homeostasis by mediating zinc uptake when cells are challenged by a zinc-restricted environment. The last study aimed to experimentally validate the presence of ZIP10 genetic variants in the human population, particularly those that might experience higher risks of zinc deficiency and anemia. We identified polymorphic ZIP10 SNPs, which were all limited to the African population, using public databases. The presence of each ZIP10 experimentally proved polymorphism with minor allele frequencies above 1% in the African population by qPCR-based genotyping. The potential implications of the African-centric ZIP10 SNPs on nutritional immunity, an innate immune response of the host organism against parasitic infections by restricting essential minerals from an invading pathogen, and precision nutrition, an approach to develop an effective diet intervention based on individual background information, were discussed in this study. In conclusion, the present studies reveal the requirement and roles of zinc and the zinc importer, ZIP10, during red blood cell development and identify the presence of African-centric ZIP10 SNPs that potentially have functional changes. The current studies suggest that restricted zinc supply and loss of ZIP10 function may play as risk factors for erythrocyte-related diseases and thus would provide an understanding of precision nutrition regarding recommendations for adequate zinc intake based on individual genetic makeup and nutritional immunity. Future researches are needed via overexpression studies in HEK293T cells, a prototypic human cell line, for functional analysis of ZIP10 SNPs to investigate their implications on zinc import. Additionally, an erythroid-specific- Zip10-floxed mice model would provide novel insight on the impact of a loss of erythroid ZIP10 at the organismal level. 145 List of references 1. Kraemer K, Zimmermann MB. Nutritional anemia. Basel: Sight and Life Press; 2007. 2. Gibson RS, Abebe Y, Stabler S, Allen RH, Westcott JE, Stoecker BJ, Krebs NF, Hambidge KM. Zinc, Gravida, Infection, and Iron, but Not Vitamin B-12 or Folate Status, Predict Hemoglobin during Pregnancy in Southern Ethiopia. The Journal of Nutrition 2008;138:581–6. 3. Siyame EWP, Hurst R, Wawer AA, Young SD, Broadley MR, Chilimba ADC, Ander LE, Watts MJ, Chilima B, Gondwe J, et al. A High Prevalence of Zinc- but not Iron-Deficiency among Women in Rural Malawi: a Cross-Sectional Study. International Journal for Vitamin and Nutrition Research 2013;83:176–87. 4. Houghton LA, Parnell WR, Thomson CD, Green TJ, Gibson RS. Serum Zinc Is a Major Predictor of Anemia and Mediates the Effect of Selenium on Hemoglobin in School-Aged Children in a Nationally Representative Survey in New Zealand. J Nutr 2016;146:1670–6. 5. Atasoy HI, Bugdayci G. Zinc deficiency and its predictive capacity for anemia: Unique model in school children. Pediatrics International 2018;60:703–9. 6. Huber KL, Cousins RJ. Zinc metabolism and metallothionein expression in bone marrow during erythropoiesis. American Journal of Physiology-Endocrinology and Metabolism 1993;264:E770–5. 7. O’Dell BL. Role of Zinc in Plasma Membrane Function. The Journal of Nutrition 2000;130:1432S-1436S. 8. Tanimura N, Liao R, Wilson GM, Dent MR, Cao M, Burstyn JN, Hematti P, Liu X, Zhang Y, Zheng Y, et al. GATA/Heme Multi-omics Reveals a Trace Metal-Dependent Cellular Differentiation Mechanism. Developmental Cell 2018;46:581-594.e4. 9. Roohani N, Hurrell R, Kelishadi R, Schulin R. Zinc and its importance for human health: An integrative review. J Res Med Sci 2013;18:144–57. 10. Kambe T, Tsuji T, Hashimoto A, Itsumura N. The Physiological, Biochemical, and Molecular Roles of Zinc Transporters in Zinc Homeostasis and Metabolism. Physiological Reviews 2015;95:749–84. 11. Katsumura KR, Bresnick EH, the GATA Factor Mechanisms Group. The GATA factor revolution in hematology. Blood 2017;129:2092–102. 12. Siatecka M, Bieker JJ. The multifunctional role of EKLF/KLF1 during erythropoiesis. Blood 2011;118:2044–54. 13. Fukai T, Ushio-Fukai M. Superoxide Dismutases: Role in Redox Signaling, Vascular Function, and Diseases. Antioxidants & Redox Signaling 2011;15:1583–606. 14. Andreini C, Bertini I. A bioinformatics view of zinc enzymes. Journal of Inorganic 146 Biochemistry 2012;111:150–6. 15. Supuran CT. Structure and function of carbonic anhydrases. Biochemical Journal 2016;473:2023–32. 16. Jaffe EK. The Remarkable Character of Porphobilinogen Synthase. Acc Chem Res 2016;49:2509–17. 17. Miyai T, Hojyo S, Ikawa T, Kawamura M, Irie T, Ogura H, Hijikata A, Bin B-H, Yasuda T, Kitamura H, et al. Zinc transporter SLC39A10/ZIP10 facilitates antiapoptotic signaling during early B-cell development. Proceedings of the National Academy of Sciences 2014;111:11780–5. 18. Tamaki M, Fujitani Y, Hara A, Uchida T, Tamura Y, Takeno K, Kawaguchi M, Watanabe T, Ogihara T, Fukunaka A, et al. The diabetes-susceptible gene SLC30A8/ZnT8 regulates hepatic insulin clearance. J Clin Invest 2013;123:4513–24. 19. Dwivedi OP, Lehtovirta M, Hastoy B, Chandra V, Krentz NAJ, Kleiner S, Jain D, Richard A-M, Abaitua F, Beer NL, et al. Loss of ZnT8 function protects against diabetes by enhanced insulin secretion. Nat Genet 2019;51:1596–606. 20. Caulfield LE, Black RE. Chapter 5. Zinc deficiency. 1st ed. World Health Org; 2004. p. 257– 80. 21. Prasad AS, Halsted JA, Nadimi M. Syndrome of iron deficiency anemia, hepatosplenomegaly, hypogonadism, dwarfism and geophagia. The American Journal of Medicine 1961;31:532–46. 22. Prasad AS, Miale A, Farid Z, Sandstead HH, Schulert AR. Zinc metabolism in patients with the syndrome of iron deficiency anemia, hepatosplenomegaly, dwarfism, and hypognadism. J Lab Clin Med 1963;61:537–49. 23. Golden MH. Specific deficiencies versus growth failure: type I and type II nutrients. SCN News 1995;10–4. 24. King JC. Zinc: an essential but elusive nutrient. The American Journal of Clinical Nutrition 2011;94:679S-684S. 25. Bakan R, Akar N. Anorexia and zinc. The Lancet 1984;324:874. 26. Pisano M, Hilas O. Zinc and Taste Disturbances in Older Adults: A Review of the Literature. Consult Pharm 2016;31:267–70. 27. Jung A, Spira D, Steinhagen-Thiessen E, Demuth I, Norman K. Zinc Deficiency Is associated With Depressive Symptoms—Results From the Berlin Aging Study II. GERONA 2016;1149–54. 28. Kawamura T, Ogawa Y, Nakamura Y, Nakamizo S, Ohta Y, Nakano H, Kabashima K, Katayama I, Koizumi S, Kodama T, et al. Severe dermatitis with loss of epidermal 147 Langerhans cells in human and mouse zinc deficiency. J Clin Invest 2012;122:722–32. 29. Gray NA, Dhana A, Stein DJ, Khumalo NP. Zinc and atopic dermatitis: a systematic review and meta‐analysis. J Eur Acad Dermatol Venereol 2019;33:1042–50. 30. Black RE, Victora CG, Walker SP, Bhutta ZA, Christian P, de Onis M, Ezzati M, Grantham- McGregor S, Katz J, Martorell R, et al. Maternal and child undernutrition and overweight in low-income and middle-income countries. The Lancet 2013;382:427–51. 31. Fischer Walker CL, Ezzati M, Black RE. Global and regional child mortality and burden of disease attributable to zinc deficiency. Eur J Clin Nutr 2009;63:591–7. 32. Prasad AS. Impact of the Discovery of Human Zinc Deficiency on Health. Journal of the American College of Nutrition 2009;28:257–65. 33. Apgar J. Zinc and Reproduction. Annu Rev Nutr 1985;5:43–68. 34. Kelkitli E, Ozturk N, Aslan NA, Kilic-Baygutalp N, Bayraktutan Z, Kurt N, Bakan N, Bakan E. Serum zinc levels in patients with iron deficiency anemia and its association with symptoms of iron deficiency anemia. Ann Hematol 2016;95:751–6. 35. O’Dell BL, Browning JD, Reeves PG. Zinc Deficiency Increases the Osmotic Fragility of Rat Erythrocytes. The Journal of Nutrition 1987;117:1883–9. 36. Johanning GL, Browning JD, Bobilya DJ, Veum TL, O’Dell BL. Effect of zinc deficiency and food restriction in the pig on erythrocyte fragility and plasma membrane composition. Nutrition Research 1990;10:1463–71. 37. Institute of Medicine (U.S.), Panel on Micronutrients. DRI: dietary reference intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium, and zinc. National Academy Press; 2002. 38. Fosmire GJ. Zinc toxicity. The American Journal of Clinical Nutrition 1990;51:225–7. 39. Sandstead HH, Freeland-Graves JH. Dietary phytate, zinc and hidden zinc deficiency. Journal of Trace Elements in Medicine and Biology 2014;28:414–7. 40. International Zinc Nutrition Consultative Group (IZiNCG), Brown KH, Rivera JA, Bhutta Z, Gibson RS, King JC, Lönnerdal B, Ruel MT, Sandtröm B, Wasantwisut E, et al. International Zinc Nutrition Consultative Group (IZiNCG) technical document #1. Assessment of the risk of zinc deficiency in populations and options for its control. Food Nutr Bull 2004;25:S99-203. 41. Institute of Medicine (U.S.) Panel on Micronutrients. Dietary reference intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium, and zinc. Washington, D.C.: National Academy Press; 2002. 42. Haase H, Overbeck S, Rink L. Zinc supplementation for the treatment or prevention of disease: Current status and future perspectives. Experimental Gerontology 2008;43:394– 148 408. 43. Igic PG, Lee E, Harper W, Roach KW. Toxic Effects Associated With Consumption of Zinc. Mayo Clinic Proceedings 2002;77:713–6. 44. Simon SR, Branda RF, Tindle BH, Burns SL. Copper deficiency and sideroblastic anemia associated with zinc ingestion. Am J Hematol 1988;28:181–3. 45. Hedera P, Peltier A, Fink JK, Wilcock S, London Z, Brewer GJ. Myelopolyneuropathy and pancytopenia due to copper deficiency and high zinc levels of unknown origin II. The denture cream is a primary source of excessive zinc. NeuroToxicology 2009;30:996–9. 46. Prasad R, Hawthorne B, Durai D, McDowell I. Zinc in denture adhesive: a rare cause of copper deficiency in a patient on home parenteral nutrition: Figure 1. BMJ Case Reports 2015;bcr2015211390. 47. Wang C-Y, Jenkitkasemwong S, Duarte S, Sparkman BK, Shawki A, Mackenzie B, Knutson MD. ZIP8 is an iron and zinc transporter whose cell-surface expression is up-regulated by cellular iron loading. J Biol Chem 2012;287:34032–43. 48. Tuschl K, Meyer E, Valdivia LE, Zhao N, Dadswell C, Abdul-Sada A, Hung CY, Simpson MA, Chong WK, Jacques TS, et al. Mutations in SLC39A14 disrupt manganese homeostasis and cause childhood-onset parkinsonism–dystonia. Nat Commun 2016;7:11601. 49. Liuzzi JP, Aydemir F, Nam H, Knutson MD, Cousins RJ. Zip14 (Slc39a14) mediates non- transferrin-bound iron uptake into cells. Proceedings of the National Academy of Sciences 2006;103:13612–7. 50. Bafaro E, Liu Y, Xu Y, Dempski RE. The emerging role of zinc transporters in cellular homeostasis and cancer. Sig Transduct Target Ther 2017;2:17029. 51. Cousins RJ. Gastrointestinal Factors Influencing Zinc Absorption and Homeostasis. International Journal for Vitamin and Nutrition Research 2010;80:243–8. 52. Lichten LA, Ryu M-S, Guo L, Embury J, Cousins RJ. MTF-1-Mediated Repression of the Zinc Transporter Zip10 Is Alleviated by Zinc Restriction. Bryk M, editor. PLoS ONE 2011;6:e21526. 53. Laity JH, Andrews GK. Understanding the mechanisms of zinc-sensing by metal-response element binding transcription factor-1 (MTF-1). Archives of Biochemistry and Biophysics 2007;463:201–10. 54. Rahman MT, Karim MM. Metallothionein: a Potential Link in the Regulation of Zinc in Nutritional Immunity. Biol Trace Elem Res 2018;182:1–13. 55. Jackson MJ. Physiology of Zinc: General Aspects. In: Mills CF, editor. Zinc in Human Biology. [Internet] London: Springer London; 1989 [cited 2020 May 17]. p. 1–14. Available from: http://link.springer.com/10.1007/978-1-4471-3879-2_1 149 56. King JC, Shames DM, Woodhouse LR. Zinc Homeostasis in Humans. The Journal of Nutrition 2000;130:1360S-1366S. 57. Krebs NF. Update on Zinc Deficiency and Excess in Clinical Pediatric Practice. Ann Nutr Metab 2013;62:19–29. 58. Taylor CM, Bacon JR, Aggett PJ, Bremner I. Homeostatic regulation of zinc absorption and endogenous losses in zinc-deprived men. The American Journal of Clinical Nutrition 1991;53:755–63. 59. Dufner-Beattie J, Wang F, Kuo Y-M, Gitschier J, Eide D, Andrews GK. The Acrodermatitis Enteropathica Gene ZIP4 Encodes a Tissue-specific, Zinc-regulated Zinc Transporter in Mice. J Biol Chem 2003;278:33474–81. 60. McMahon RJ, Cousins RJ. Regulation of the zinc transporter ZnT-1 by dietary zinc. Proceedings of the National Academy of Sciences 1998;95:4841–6. 61. Liuzzi JP, Blanchard RK, Cousins RJ. Differential Regulation of Zinc Transporter 1, 2, and 4 mRNA Expression by Dietary Zinc in Rats. The Journal of Nutrition 2001;131:46–52. 62. Dufner-Beattie J, Kuo Y-M, Gitschier J, Andrews GK. The Adaptive Response to Dietary Zinc in Mice Involves the Differential Cellular Localization and Zinc Regulation of the Zinc Transporters ZIP4 and ZIP5. J Biol Chem 2004;279:49082–90. 63. Guo L, Cousins RJ. ZnT1 (SLC30A1) and ZnT2 (SLC30A2) regulate zinc efflux through two different pathways in pancreatic acinar cells. FASEB j 2008;22:674–674. 64. Guo L, Lichten LA, Ryu M-S, Liuzzi JP, Wang F, Cousins RJ. STAT5-glucocorticoid receptor interaction and MTF-1 regulate the expression of ZnT2 (Slc30a2) in pancreatic acinar cells. Proceedings of the National Academy of Sciences 2010;107:2818–23. 65. Gardner W, Kassebaum N. Global, Regional, and National Prevalence of Anemia and Its Causes in 204 Countries and Territories, 1990–2019. Current Developments in Nutrition 2020;4:830–830. 66. Gardner W, Kassebaum N. Global, Regional, and National Prevalence of Anemia and Its Causes in 204 Countries and Territories, 1990–2019. Current Developments in Nutrition 2020;4:830–830. 67. Alarcon K, Kolsteren PW, Prada AM, Chian AM, Velarde RE, Pecho IL, Hoeree TF. Effects of separate delivery of zinc or zinc and vitamin A on hemoglobin response, growth, and diarrhea in young Peruvian children receiving iron therapy for anemia. The American Journal of Clinical Nutrition 2004;80:1276–82. 68. Chen Y-H, Jeng S-S, Hsu Y-C, Liao Y-M, Wang Y-X, Cao X, Huang L-J. In anemia zinc is recruited from bone and plasma to produce new red blood cells. Journal of Inorganic Biochemistry 2020;210:111172. 69. Taysi S, Cikman O, Kaya A, Demircan B, Gumustekin K, Yilmaz A, Boyuk A, Keles M, 150 Akyuz M, Turkeli M. Increased Oxidant Stress and Decreased Antioxidant Status in Erythrocytes of Rats Fed with Zinc-deficient Diet. Biol Trace Elem Res 2008;123:161–7. 70. Kraus A, Roth H-P, Kirchgessner M. Supplementation with Vitamin C, Vitamin E or β- Carotene Influences Osmotic Fragility and Oxidative Damage of Erythrocytes of Zinc- Deficient Rats. The Journal of Nutrition 1997;127:1290–6. 71. Konomi A, Yokoi K. Zinc Deficiency Decreases Plasma Erythropoietin Concentration in Rats. BTER 2005;107:289–92. 72. Koury S, Yarlagadda S, Moskalik-Liermo K, Popli N, Kim N, Apolito C, Peterson A, Zhang X, Zu P, Tamburlin J, et al. Differential gene expression during terminal erythroid differentiation. Genomics 2007;90:574–82. 73. Tallack MR, Keys JR, Humbert PO, Perkins AC. EKLF/KLF1 Controls Cell Cycle Entry via Direct Regulation of E2f2. Journal of Biological Chemistry 2009;284:20966–74. 74. Gutiérrez L, Caballero N, Fernández‐Calleja L, Karkoulia E, Strouboulis J. Regulation of GATA1 levels in erythropoiesis. IUBMB Life 2020;72:89–105. 75. Briegel K, Bartunek P, Stengl G, Lim K-C, Beug H, Engel JD, Zenke M. Regulation and function of transcription factor GATA-1 during red blood cell differentiation. Development 1996;122:3839–50. 76. Ferreira R, Ohneda K, Yamamoto M, Philipsen S. GATA1 Function, a Paradigm for Transcription Factors in Hematopoiesis. MCB 2005;25:1215–27. 77. Moriguchi T, Yamamoto M. A regulatory network governing Gata1 and Gata2 gene transcription orchestrates erythroid lineage differentiation. Int J Hematol 2014;100:417–24. 78. Krishnamurthy VM, Kaufman GK, Urbach AR, Gitlin I, Gudiksen KL, Weibel DB, Whitesides GM. Carbonic Anhydrase as a Model for Biophysical and Physical-Organic Studies of Proteins and Protein−Ligand Binding. Chem Rev 2008;108:946–1051. 79. Imtaiyaz Hassan Md, Shajee B, Waheed A, Ahmad F, Sly WS. Structure, function and applications of carbonic anhydrase isozymes. Bioorganic & Medicinal Chemistry 2013;21:1570–82. 80. Rishi G, Subramaniam VN. The relationship between systemic iron homeostasis and erythropoiesis. Bioscience Reports 2017;37:BSR20170195. 81. Zhang DL, Wu J, Shah BN, Greutélaers KC, Ghosh MC, Ollivierre H, Su XZ, Thuma PE, Bedu-Addo G, Mockenhaupt FP, et al. Erythrocytic ferroportin reduces intracellular iron accumulation, hemolysis, and malaria risk. Science 2018;359:1520–3. 82. Fujiwara T, Harigae H. Biology of Heme in Mammalian Erythroid Cells and Related Disorders. BioMed Research International 2015;2015:1–9. 83. Chau TT, Ishigaki M, Kataoka T, Taketani S. Ferrochelatase Catalyzes the Formation of Zn- 151 protoporphyrin of Dry-Cured Ham via the Conversion Reaction from Heme in Meat. J Agric Food Chem 2011;59:12238–45. 84. Rettmer RL, Carlson TH, Origenes ML, Md J, Jack RM, Labbé RF. Zinc Protoporphyrin/Heme Ratio for Diagnosis of Preanemic Iron Deficiency. Pediatrics 1999;104:e37–e37. 85. Ryu M-S, Lichten LA, Liuzzi JP, Cousins RJ. Zinc Transporters ZnT1 (Slc30a1), Zip8 (Slc39a8), and Zip10 (Slc39a10) in Mouse Red Blood Cells Are Differentially Regulated during Erythroid Development and by Dietary Zinc Deficiency. The Journal of Nutrition 2008;138:2076–83. 86. Chen J, Gálvez-Peralta M, Zhang X, Deng J, Liu Z, Nebert DW. In utero gene expression in the Slc39a8(neo/neo) knockdown mouse. Sci Rep 2018;8:10703. 87. Kemna E, Pickkers P, Nemeth E, van der Hoeven H, Swinkels D. Time-course analysis of hepcidin, serum iron, and plasma cytokine levels in humans injected with LPS. Blood 2005;106:1864–6. 88. Gaetke LM, McClain CJ, Talwalkar RT, Shedlofsky SI. Effects of endotoxin on zinc metabolism in human volunteers. American Journal of Physiology-Endocrinology and Metabolism 1997;272:E952–6. 89. Beker Aydemir T, Chang S-M, Guthrie GJ, Maki AB, Ryu M-S, Karabiyik A, Cousins RJ. Zinc Transporter ZIP14 Functions in Hepatic Zinc, Iron and Glucose Homeostasis during the Innate Immune Response (Endotoxemia). Nerurkar PV, editor. PLoS ONE 2012;7:e48679. 90. Pigeon C, Ilyin G, Courselaud B, Leroyer P, Turlin B, Brissot P, Loréal O. A New Mouse Liver-specific Gene, Encoding a Protein Homologous to Human Antimicrobial Peptide Hepcidin, Is Overexpressed during Iron Overload. J Biol Chem 2001;276:7811–9. 91. Wrighting DM, Andrews NC. Interleukin-6 induces hepcidin expression through STAT3. Blood 2006;108:3204–9. 92. Nemeth E, Valore EV, Territo M, Schiller G, Lichtenstein A, Ganz T. Hepcidin, a putative mediator of anemia of inflammation, is a type II acute-phase protein. Blood 2003;101:2461– 3. 93. Khan FA, Fisher MA, Khakoo RA. Association of hemochromatosis with infectious diseases: expanding spectrum. International Journal of Infectious Diseases 2007;11:482–7. 94. Weinberg ED. Iron availability and infection. Biochimica et Biophysica Acta (BBA) - General Subjects 2009;1790:600–5. 95. Kehl-Fie TE, Skaar EP. Nutritional immunity beyond iron: a role for manganese and zinc. Current Opinion in Chemical Biology 2010;14:218–24. 96. Liuzzi JP, Lichten LA, Rivera S, Blanchard RK, Aydemir TB, Knutson MD, Ganz T, Cousins 152 RJ. Interleukin-6 regulates the zinc transporter Zip14 in liver and contributes to the hypozincemia of the acute-phase response. Proceedings of the National Academy of Sciences 2005;102:6843–8. 97. Corbin BD, Seeley EH, Raab A, Feldmann J, Miller MR, Torres VJ, Anderson KL, Dattilo BM, Dunman PM, Gerads R, et al. Metal Chelation and Inhibition of Bacterial Growth in Tissue Abscesses. Science 2008;319:962–5. 98. Wang K, Zhou B, Kuo Y-M, Zemansky J, Gitschier J. A Novel Member of a Zinc Transporter Family Is Defective in Acrodermatitis Enteropathica. The American Journal of Human Genetics 2002;71:66–73. 99. Drake I, Hindy G, Ericson U, Orho-Melander M. A prospective study of dietary and supplemental zinc intake and risk of type 2 diabetes depending on genetic variation in SLC30A8. Genes Nutr 2017;12:30. 100. Bin B, Hojyo S, Hosaka T, Bhin J, Kano H, Miyai T, Ikeda M, Kimura‐Someya T, Shirouzu M, Cho E, et al. Molecular pathogenesis of Spondylocheirodysplastic Ehlers‐Danlos syndrome caused by mutant ZIP13 proteins. EMBO Mol Med 2014;6:1028–42. 101. Anzilotti C, Swan DJ, Boisson B, Deobagkar-Lele M, Oliveira C, Chabosseau P, Engelhardt KR, Xu X, Chen R, Alvarez L, et al. An essential role for the Zn2+ transporter ZIP7 in B cell development. Nat Immunol 2019;20:350–61. 102. Chowanadisai W, Lönnerdal B, Kelleher SL. Identification of a Mutation in SLC30A2 (ZnT- 2) in Women with Low Milk Zinc Concentration That Results in Transient Neonatal Zinc Deficiency. J Biol Chem 2006;281:39699–707. 103. Quadri M, Federico A, Zhao T, Breedveld GJ, Battisti C, Delnooz C, Severijnen L-A, Di Toro Mammarella L, Mignarri A, Monti L, et al. Mutations in SLC30A10 Cause Parkinsonism and Dystonia with Hypermanganesemia, Polycythemia, and Chronic Liver Disease. The American Journal of Human Genetics 2012;90:467–77. 104. Mbacham WF, Ayong L, Guewo-Fokeng M, Makoge V. Current Situation of Malaria in Africa. In: Ariey F, Gay F, Ménard R, editors. Malaria Control and Elimination. Springer New York; 2019. p. 29–44. 105. Kasvosve I, Gomo ZA, Nathoo KJ, Matibe P, Mudenge B, Loyevsky M, Gordeuk VR. Effect of ferroportin Q248H polymorphism on iron status in African children. The American Journal of Clinical Nutrition 2005;82:1102–6. 106. Engelken J, Carnero-Montoro E, Pybus M, Andrews GK, Lalueza-Fox C, Comas D, Sekler I, de la Rasilla M, Rosas A, Stoneking M, et al. Extreme Population Differences in the Human Zinc Transporter ZIP4 (SLC39A4) Are Explained by Positive Selection in Sub- Saharan Africa. Akey JM, editor. PLoS Genet 2014;10:e1004128. 107. Marvin RG, Wolford JL, Kidd MJ, Murphy S, Ward J, Que EL, Mayer ML, Penner-Hahn JE, Haldar K, O’Halloran TV. Fluxes in “Free” and Total Zinc Are Essential for Progression of Intraerythrocytic Stages of Plasmodium falciparum. Chemistry & Biology 2012;19:731– 153 41. 108. Niles JC. Malarial Parasites Accumulate Labile Zinc Pools. Chemistry & Biology 2012;19:660–1. 109. Omichinski J, Clore G, Schaad O, Felsenfeld G, Trainor C, Appella E, Stahl S, Gronenborn A. NMR structure of a specific DNA complex of Zn-containing DNA binding domain of GATA-1. Science 1993;261:438–46. 110. Quadrini KJ, Bieker JJ. Krüppel-like Zinc Fingers Bind to Nuclear Import Proteins and Are Required for Efficient Nuclear Localization of Erythroid Krüppel-like Factor. J Biol Chem 2002;277:32243–52. 111. Dent AJ, Beyersmann D, Block C, Hasnain SS. Two different zinc sites in bovine 5- aminolevulinate dehydratase distinguished by extended x-ray absorption fine structure. Biochemistry 1990;29:7822–8. 112. Jaffe EK, Abrams WR, Kaempfen HX, Harris KA. 5-Chlorolevulinate modification of porphobilinogen synthase identifies a potential role for the catalytic zinc. Biochemistry 1992;31:2113–23. 113. Muckenthaler MU, Rivella S, Hentze MW, Galy B. A Red Carpet for Iron Metabolism. Cell 2017;168:344–61. 114. Taketani S, Tokunaga R. Purification and Substrate Specificity of Bovine Liver- Ferrochelatase. European Journal of Biochemistry 2005;127:443–7. 115. Bloomer JR, Reuter RJ, Morton KO, Wehner JM. Enzymatic formation of zinc- protoporphyrin by rat liver and its potential effect on hepatic heme metabolism. Gastroenterology 1983;85:663–8. 116. Jain D, Mishra T, Giardine BM, Keller CA, Morrissey CS, Magargee S, Dorman CM, Long M, Weiss MJ, Hardison RC. Dynamics of GATA1 binding and expression response in a GATA1-induced erythroid differentiation system. Genomics Data 2015;4:1–7. 117. Grillo AS, SantaMaria AM, Kafina MD, Cioffi AG, Huston NC, Han M, Seo YA, Yien YY, Nardone C, Menon AV, et al. Restored iron transport by a small molecule promotes absorption and hemoglobinization in animals. Science 2017;356:608–16. 118. Cho Y-E, Lomeda R-AR, Ryu S-H, Lee J-H, Beattie JH, Kwun I-S. Cellular Zn depletion by metal ion chelators (TPEN, DTPA and chelex resin) and its application to osteoblastic MC3T3-E1 cells. Nutr Res Pract 2007;1:29. 119. Goedhart J, Luijsterburg MS. VolcaNoseR is a web app for creating, exploring, labeling and sharing volcano plots. Sci Rep 2020;10:20560. 120. Krämer A, Green J, Pollard J, Tugendreich S. Causal analysis approaches in Ingenuity Pathway Analysis. Bioinformatics 2014;30:523–30. 154 121. Ryu M-S, Zhang D, Protchenko O, Shakoury-Elizeh M, Philpott CC. PCBP1 and NCOA4 regulate erythroid iron storage and heme biosynthesis. Journal of Clinical Investigation 2017;127:1786–97. 122. Paradkar PN, Zumbrennen KB, Paw BH, Ward DM, Kaplan J. Regulation of Mitochondrial Iron Import through Differential Turnover of Mitoferrin 1 and Mitoferrin 2. MCB 2009;29:1007–16. 123. Troadec M-B, Warner D, Wallace J, Thomas K, Spangrude GJ, Phillips J, Khalimonchuk O, Paw BH, Ward DM, Kaplan J. Targeted deletion of the mouse Mitoferrin1 gene: from anemia to protoporphyria. Blood 2011;117:5494–502. 124. Parsons PJ. Measurement of Erythrocyte Protoporphyrin Concentration by Double Extraction and Spectrofluorometry. Current Protocols in Toxicology [Internet] 1999 [cited 2021 Oct 30];2. Available from: https://onlinelibrary.wiley.com/doi/10.1002/0471140856.tx0808s02 125. Ganz T, Nemeth E. Iron Metabolism: Interactions with Normal and Disordered Erythropoiesis. Cold Spring Harbor Perspectives in Medicine 2012;2:a011668–a011668. 126. Mendivil-Perez M, Velez-Pardo C, Jimenez-Del-Rio M. TPEN Induces Apoptosis Independently of Zinc Chelator Activity in a Model of Acute Lymphoblastic Leukemia and Ex Vivo Acute Leukemia Cells through Oxidative Stress and Mitochondria Caspase-3- and AIF-Dependent Pathways. Oxidative Medicine and Cellular Longevity 2012;2012:1–14. 127. Gao H, Zhao L, Wang H, Xie E, Wang X, Wu Q, Yu Y, He X, Ji H, Rink L, et al. Metal transporter Slc39a10 regulates susceptibility to inflammatory stimuli by controlling macrophage survival. Proc Natl Acad Sci USA 2017;114:12940–5. 128. Tahara T, Sun J, Igarashi K, Taketani S. Heme-dependent up-regulation of the α-globin gene expression by transcriptional repressor Bach1 in erythroid cells. Biochemical and Biophysical Research Communications 2004;324:77–85. 129. Chen J-J. Regulation of protein synthesis by the heme-regulated eIF2α kinase: relevance to anemias. Blood 2007;109:2693–9. 130. Fillebeen C, Charlebois E, Wagner J, Katsarou A, Mui J, Vali H, Garcia-Santos D, Ponka P, Presley J, Pantopoulos K. Transferrin receptor 1 controls systemic iron homeostasis by fine- tuning hepcidin expression to hepatocellular iron load. Blood 2019;133:344–55. 131. Wang J, Chen G, Muckenthaler M, Galy B, Hentze MW, Pantopoulos K. Iron-Mediated Degradation of IRP2, an Unexpected Pathway Involving a 2-Oxoglutarate-Dependent Oxygenase Activity. Mol Cell Biol 2004;24:954–65. 132. Welch JJ. Global regulation of erythroid gene expression by transcription factor GATA-1. Blood 2004;104:3136–47. 133. Hastka J, Lasserre JJ, Schwarzbeck A, Hehlmann R. Central role of zinc protoporphyrin in staging iron deficiency. Clin Chem 1994;40:768–73. 155 134. Kumssa DB, Joy EJM, Ander EL, Watts MJ, Young SD, Walker S, Broadley MR. Dietary calcium and zinc deficiency risks are decreasing but remain prevalent. Sci Rep 2015;5:10974. 135. Killilea DW, Rohner F, Ghosh S, Otoo GE, Smith L, Siekmann JH, King JC. Identification of a Hemolysis Threshold That Increases Plasma and Serum Zinc Concentration. J Nutr 2017;147:1218–25. 136. Faraji B, Swendseid ME. Growth Rate, Tissue Zinc Levels and Activities of Selected Enzymes in Rats Fed a Zinc-Deficient Diet by Gastric Tube. The Journal of Nutrition 1983;113:447–55. 137. Abdulla M, Svensson S. Effect of oral zinc intake on δ-aminolaevulinic acid dehydratase in red blood cells. Scandinavian Journal of Clinical and Laboratory Investigation 1979;39:31– 6. 138. Hodge D, Coghill E, Keys J, Maguire T, Hartmann B, McDowall A, Weiss M, Grimmond S, Perkins A. A global role for EKLF in definitive and primitive erythropoiesis. Blood 2006;107:3359–70. 139. Lutton JD, Abraham NG, Drummond GS, Levere RD, Kappas A. Zinc porphyrins: Potent inhibitors of hematopoieses in animal and human bone marrow. Proceedings of the National Academy of Sciences 1997;94:1432–6. 140. Thurnham D, Northrop-Clewes C. Biomarkers for the differentiation of anemia and their clinical usefulness. JBM 2013;11. 141. Joint World Health Organization/Centers for Disease Control and Prevention Technical Consultation on the Assessment of Iron Status at the Population Level (2004ௗ: Geneva S. Assessing the iron status of populations: including literature reviews: report of a Joint World Health Organization/Centers for Disease Control and Prevention Technical Consultation on the Assessment of Iron Status at the Population Level, Geneva, Switzerland, 6-8 April 2004 [Internet]. 2nd ed. Geneva: World Health Organization; 2007 [cited 2021 Dec 18]. Available from: https://apps.who.int/iris/handle/10665/75368 142. Rylski M, Welch JJ, Chen Y-Y, Letting DL, Diehl JA, Chodosh LA, Blobel GA, Weiss MJ. GATA-1-Mediated Proliferation Arrest during Erythroid Maturation. Mol Cell Biol 2003;23:5031–42. 143. Tanimura N, Miller E, Igarashi K, Yang D, Burstyn JN, Dewey CN, Bresnick EH. Mechanism governing heme synthesis reveals a GATA factor/heme circuit that controls differentiation. EMBO Rep 2016;17:249–65. 144. Jankowski-Hennig MA, Clegg MS, Daston GP, Rogers JM, Keen CL. Zinc-Deficient Rat Embryos Have Increased Caspase 3-like Activity and Apoptosis. Biochemical and Biophysical Research Communications 2000;271:250–6. 145. Marreiro D, Cruz K, Morais J, Beserra J, Severo J, de Oliveira A. Zinc and Oxidative Stress: Current Mechanisms. Antioxidants 2017;6:24. 156 146. Papanikolaou G, Pantopoulos K. Iron metabolism and toxicity. Toxicology and Applied Pharmacology 2005;202:199–211. 147. Chen X, Yu C, Kang R, Tang D. Iron Metabolism in Ferroptosis. Front Cell Dev Biol 2020;8:590226. 148. Çimen MYB. Free radical metabolism in human erythrocytes. Clinica Chimica Acta 2008;390:1–11. 149. Ebert PS, Hess RA, Frykholm BC, Tschudy DP. Succinylacetone, a potent inhibitor of heme biosynthesis: Effect on cell growth, heme content and δ-aminolevulinic acid dehydratase activity of malignant murine erythroleukemia cells. Biochemical and Biophysical Research Communications 1979;88:1382–90. 150. Moon S-H, Huang C-H, Houlihan SL, Regunath K, Freed-Pastor WA, Morris JP, Tschaharganeh DF, Kastenhuber ER, Barsotti AM, Culp-Hill R, et al. p53 Represses the Mevalonate Pathway to Mediate Tumor Suppression. Cell 2019;176:564-580.e19. 151. Zeng L, Liao H, Liu Y, Lee T-S, Zhu M, Wang X, Stemerman MB, Zhu Y, Shyy JY-J. Sterol- responsive Element-binding Protein (SREBP) 2 Down-regulates ATP-binding Cassette Transporter A1 in Vascular Endothelial Cells. Journal of Biological Chemistry 2004;279:48801–7. 152. Battaglia AM, Chirillo R, Aversa I, Sacco A, Costanzo F, Biamonte F. Ferroptosis and Cancer: Mitochondria Meet the “Iron Maiden” Cell Death. Cells 2020;9:1505. 153. Ou Y, Wang S-J, Li D, Chu B, Gu W. Activation of SAT1 engages polyamine metabolism with p53-mediated ferroptotic responses. Proc Natl Acad Sci USA 2016;113:E6806–12. 154. Garcia-Bermudez J, Baudrier L, Bayraktar EC, Shen Y, La K, Guarecuco R, Yucel B, Fiore D, Tavora B, Freinkman E, et al. Squalene accumulation in cholesterol auxotrophic lymphomas prevents oxidative cell death. Nature 2019;567:118–22. 155. Tang D, Chen X, Kang R, Kroemer G. Ferroptosis: molecular mechanisms and health implications. Cell Res 2021;31:107–25. 156. Wu J, Minikes AM, Gao M, Bian H, Li Y, Stockwell BR, Chen Z-N, Jiang X. Intercellular interaction dictates cancer cell ferroptosis via NF2–YAP signalling. Nature 2019;572:402– 6. 157. Stockwell BR, Friedmann Angeli JP, Bayir H, Bush AI, Conrad M, Dixon SJ, Fulda S, Gascón S, Hatzios SK, Kagan VE, et al. Ferroptosis: A Regulated Cell Death Nexus Linking Metabolism, Redox Biology, and Disease. Cell 2017;171:273–85. 158. Luo M, Wu L, Zhang K, Wang H, Zhang T, Gutierrez L, O’Connell D, Zhang P, Li Y, Gao T, et al. miR-137 regulates ferroptosis by targeting glutamine transporter SLC1A5 in melanoma. Cell Death Differ 2018;25:1457–72. 159. Song X, Xie Y, Kang R, Hou W, Sun X, Epperly MW, Greenberger JS, Tang D. FANCD2 157 protects against bone marrow injury from ferroptosis. Biochemical and Biophysical Research Communications 2016;480:443–9. 160. Palmiter RD, Findley SD. Cloning and functional characterization of a mammalian zinc transporter that confers resistance to zinc. The EMBO Journal 1995;14:639–49. 161. Zheng D, Feeney GP, Kille P, Hogstrand C. Regulation of ZIP and ZnT zinc transporters in zebrafish gill: zinc repression of ZIP10 transcription by an intronic MRE cluster. Physiological Genomics 2008;34:205–14. 162. Wieringa F, Dijkhuizen M, Fiorentino M, Laillou A, Berger J. Determination of Zinc Status in Humans: Which Indicator Should We Use? Nutrients 2015;7:3252–63. 163. King JC, Shames DM, Lowe NM, Woodhouse LR, Sutherland B, Abrams SA, Turnlund JR, Jackson MJ. Effect of acute zinc depletion on zinc homeostasis and plasma zinc kinetics in men. The American Journal of Clinical Nutrition 2001;74:116–24. 164. Prasad AS, Mantzoros CS, Beck FWJ, Hess JW, Brewer GJ. Zinc status and serum testosterone levels of healthy adults. Nutrition 1996;12:344–8. 165. Hininger-Favier I, Andriollo-Sanchez M, Arnaud J, Meunier N, Bord S, Graham C, Polito A, Maiani G, O’Connor JM, Coudray C, et al. Age- and sex-dependent effects of long-term zinc supplementation on essential trace element status and lipid metabolism in European subjects: the Zenith Study. Br J Nutr 2007;97:569–78. 166. Lowe NM, Fekete K, Decsi T. Methods of assessment of zinc status in humans: a systematic review. The American Journal of Clinical Nutrition 2009;89:2040S-2051S. 167. Moran VH, Stammers A-L, Medina MW, Patel S, Dykes F, Souverein OW, Dullemeijer C, Pérez-Rodrigo C, Serra-Majem L, Nissensohn M, et al. The Relationship between Zinc Intake and Serum/Plasma Zinc Concentration in Children: A Systematic Review and Dose- Response Meta-Analysis. Nutrients 2012;4:841–58. 168. Hess SY, Peerson JM, King JC, Brown KH. Use of Serum Zinc Concentration as an Indicator of Population Zinc Status. Food Nutr Bull 2007;28:S403–29. 169. Lowe NM. Assessing zinc in humans. Current Opinion in Clinical Nutrition & Metabolic Care 2016;19:321–7. 170. Hennigar SR, Lieberman HR, Fulgoni VL, McClung JP. Serum Zinc Concentrations in the US Population Are Related to Sex, Age, and Time of Blood Draw but Not Dietary or Supplemental Zinc. The Journal of Nutrition 2018;148:1341–51. 171. Workinger J, Doyle Robert, Bortz J. Challenges in the Diagnosis of Magnesium Status. Nutrients 2018;10:1202. 172. Thomas EA, Bailey LB, Kauwell GA, Lee D-Y, Cousins RJ. Erythrocyte Metallothionein Response to Dietary Zinc in Humans. The Journal of Nutrition 1992;122:2408–14. 158 173. Ruz M, Cavan KR, Bettger WJ, Gibson RS. Erythrocytes, erythrocyte membranes, neutrophils and platelets as biopsy materials for the assessment of zinc status in humans. Br J Nutr 1992;68:515–27. 174. Hojyo S, Miyai T, Fujishiro H, Kawamura M, Yasuda T, Hijikata A, Bin B-H, Irie T, Tanaka J, Atsumi T, et al. Zinc transporter SLC39A10/ZIP10 controls humoral immunity by modulating B-cell receptor signal strength. Proceedings of the National Academy of Sciences 2014;111:11786–91. 175. Bin B-H, Bhin J, Takaishi M, Toyoshima K, Kawamata S, Ito K, Hara T, Watanabe T, Irié T, Takagishi T, et al. Requirement of zinc transporter ZIP10 for epidermal development: Implication of the ZIP10–p63 axis in epithelial homeostasis. Proc Natl Acad Sci USA 2017;114:12243–8. 176. Dixon SJ, Lemberg KM, Lamprecht MR, Skouta R, Zaitsev EM, Gleason CE, Patel DN, Bauer AJ, Cantley AM, Yang WS, et al. Ferroptosis: An Iron-Dependent Form of Nonapoptotic Cell Death. Cell 2012;149:1060–72. 177. Hao S, Liang B, Huang Q, Dong S, Wu Z, He W, Shi M. Metabolic networks in ferroptosis (Review). Oncol Lett [Internet] 2018 [cited 2021 Dec 15]; Available from: http://www.spandidos-publications.com/10.3892/ol.2018.8066 178. Chen L, Ma M-Y, Sun M, Jiang L-Y, Zhao X-T, Fang X-X, Man Lam S, Shui G-H, Luo J, Shi X-J, et al. Endogenous sterol intermediates of the mevalonate pathway regulate HMGCR degradation and SREBP-2 processing. Journal of Lipid Research 2019;60:1765– 75. 179. Hall CL, Sutanto H, Dalageorgou C, McKenna WJ, Syrris P, Futema M. Frequency of genetic variants associated with arrhythmogenic right ventricular cardiomyopathy in the genome aggregation database. Eur J Hum Genet 2018;26:1312–8. 180. Rentzsch P, Witten D, Cooper GM, Shendure J, Kircher M. CADD: predicting the deleteriousness of variants throughout the human genome. Nucleic Acids Research 2019;47:D886–94. 181. Marcus JH, Novembre J. Visualizing the geography of genetic variants. Bioinformatics 2016;btw643. 182. Nugent T, Ward S, Jones DT. The MEMPACK alpha-helical transmembrane protein structure prediction server. Bioinformatics 2011;27:1438–9. 183. Cusick SE, Jaramillo EG, Moody EC, Ssemata AS, Bitwayi D, Lund TC, Mupere E. Assessment of blood levels of heavy metals including lead and manganese in healthy children living in the Katanga settlement of Kampala, Uganda. BMC Public Health 2018;18:717. 184. Taylor KM, Morgan HE, Smart K, Zahari NM, Pumford S, Ellis IO, Robertson JFR, Nicholson RI. The Emerging Role of the LIV-1 Subfamily of Zinc Transporters in Breast Cancer. Mol Med 2007;13:396–406. 159 185. Taylor KM, Nicholson RI. The LZT proteins; the LIV-1 subfamily of zinc transporters. Biochimica et Biophysica Acta (BBA) - Biomembranes 2003;1611:16–30. 186. Taylor KM, Muraina IA, Brethour D, Schmitt-Ulms G, Nimmanon T, Ziliotto S, Kille P, Hogstrand C. Zinc transporter ZIP10 forms a heteromer with ZIP6 which regulates embryonic development and cell migration. Biochemical Journal 2016;473:2531–44. 187. Lichten LA, Cousins RJ. Mammalian Zinc Transporters: Nutritional and Physiologic Regulation. Annu Rev Nutr 2009;29:153–76. 188. Guerinot ML. The ZIP family of metal transporters. Biochimica et Biophysica Acta (BBA) - Biomembranes 2000;1465:190–8. 189. Fukada T, Kambe T. Molecular and genetic features of zinc transporters in physiology and pathogenesis. Metallomics 2011;3:662. 190. Dash R, Mitra S, Munni YA, Choi HJ, Ali MdC, Barua L, Jang TJ, Moon IS. Computational Insights into the Deleterious Impacts of Missense Variants on N-Acetyl-d-glucosamine Kinase Structure and Function. IJMS 2021;22:8048. 191. George DCP, Chakraborty C, Haneef SS, NagaSundaram N, Chen L, Zhu H. Evolution- and Structure-Based Computational Strategy Reveals the Impact of Deleterious Missense Mutations on MODY 2 (Maturity-Onset Diabetes of the Young, Type 2). Theranostics 2014;4:366–85. 192. Berhe K, Gebrearegay F, Gebremariam H. Prevalence and associated factors of zinc deficiency among pregnant women and children in Ethiopia: a systematic review and meta- analysis. BMC Public Health 2019;19:1663. 193. Gammoh NZ, Rink L. Zinc in Infection and Inflammation. Nutrients 2017;9. 194. Hood MI, Skaar EP. Nutritional immunity: transition metals at the pathogen–host interface. Nat Rev Microbiol 2012;10:525–37. 195. Subramanian Vignesh K, Landero Figueroa JA, Porollo A, Caruso JA, Deepe GS. Granulocyte Macrophage-Colony Stimulating Factor Induced Zn Sequestration Enhances Macrophage Superoxide and Limits Intracellular Pathogen Survival. Immunity 2013;39:697–710. 196. Nolin E, Gans S, Llamas L, Bandyopadhyay S, Brittain SM, Bernasconi-Elias P, Carter KP, Loureiro JJ, Thomas JR, Schirle M, et al. Discovery of a ZIP7 inhibitor from a Notch pathway screen. Nat Chem Biol 2019;15:179–88. 197. Go-T2D Consortium, T2D-GENES Consortium, Flannick J, Thorleifsson G, Beer NL, Jacobs SBR, Grarup N, Burtt NP, Mahajan A, Fuchsberger C, et al. Loss-of-function mutations in SLC30A8 protect against type 2 diabetes. Nat Genet 2014;46:357–63. 198. Merriman C, Li H, Li H, Fu D. Highly specific monoclonal antibodies for allosteric inhibition and immunodetection of the human pancreatic zinc transporter ZnT8. Journal of 160 Biological Chemistry 2018;293:16206–16. 199. Pujol‐Giménez J, Poirier M, Bühlmann S, Schuppisser C, Bhardwaj R, Awale M, Visini R, Javor S, Hediger MA, Reymond J. Inhibitors of Human Divalent Metal Transporters DMT1 (SLC11A2) and ZIP8 (SLC39A8) from a GDB‐17 Fragment Library. ChemMedChem 2021;16:3306–14. 200. Wang F. Acrodermatitis enteropathica mutations affect transport activity, localization and zinc-responsive trafficking of the mouse ZIP4 zinc transporter. Human Molecular Genetics 2004;13:563–71. 201. Kok F, Bouwman L, Desiere F, editors. Personalized nutrition: principles and applications. Boca Raton: CRC Press/Taylor & Francis; 2008. 287 p. 202. Klei TRL, Meinderts SM, van den Berg TK, van Bruggen R. From the Cradle to the Grave: The Role of Macrophages in Erythropoiesis and Erythrophagocytosis. Front Immunol [Internet] 2017 [cited 2021 Dec 22];8. Available from: http://journal.frontiersin.org/article/10.3389/fimmu.2017.00073/full 203. Lin H, Burton D, Li L, Warner DE, Phillips JD, McVey Ward D, Kaplan J. Gain-of-function mutations identify amino acids within transmembrane domains of the yeast vacuolar transporter Zrc1 that determine metal specificity. Biochemical Journal 2009;422:273–83. 204. Seo YA, Kelleher SL. Functional analysis of two single nucleotide polymorphisms in SLC30A2 (ZnT2): implications for mammary gland function and breast disease in women. Physiological Genomics 2010;42A:219–27. 205. Eide D, Broderius M, Fett J, Guerinot ML. A novel iron-regulated metal transporter from plants identified by functional expression in yeast. Proceedings of the National Academy of Sciences 1996;93:5624–8. 206. Murgia C, Vespignani I, Cerase J, Nobili F, Perozzi G. Cloning, expression, and vesicular localization of zinc transporter Dri 27/ZnT4 in intestinal tissue and cells. American Journal of Physiology-Gastrointestinal and Liver Physiology 1999;277:G1231–9. 207. Gitan RS, Shababi M, Kramer M, Eide DJ. A Cytosolic Domain of the Yeast Zrt1 Zinc Transporter Is Required for Its Post-translational Inactivation in Response to Zinc and Cadmium. Journal of Biological Chemistry 2003;278:39558–64. 208. Milon B, Wu Q, Zou J, Costello LC, Franklin RB. Histidine residues in the region between transmembrane domains III and IV of hZip1 are required for zinc transport across the plasma membrane in PC-3 cells. Biochimica et Biophysica Acta (BBA) - Biomembranes 2006;1758:1696–701. 161 Appendices Appendix1. Amino acid sequences for alignments >NP_065075.1 zinc transporter ZIP10 precursor [Homo sapiens] MKVHMHTKFCLICLLTFIFHHCNHCHEEHDHGPEALHRQHRGMTELEPSKFSKQAAENE KKYYIEKLFERYGENGRLSFFGLEKLLTNLGLGERKVVEINHEDLGHDHVSHLDILAVQE GKHFHSHNHQHSHNHLNSENQTVTSVSTKRNHKCDPEKETVEVSVKSDDKHMHDHNH RLRHHHRLHHHLDHNNTHHFHNDSITPSERGEPSNEPSTETNKTQEQSDVKLPKGKRKK KGRKSNENSEVITPGFPPNHDQGEQYEHNRVHKPDRVHNPGHSHVHLPERNGHDPGRG HQDLDPDNEGELRHTRKREAPHVKNNAIISLRKDLNEDDHHHECLNVTQLLKYYGHGA NSPISTDLFTYLCPALLYQIDSRLCIEHFDKLLVEDINKDKNLVPEDEANIGASAWICGIISI TVISLLSLLGVILVPIINQGCFKFLLTFLVALAVGTMSGDALLHLLPHSQGGHDHSHQHAH GHGHSHGHESNKFLEEYDAVLKGLVALGGIYLLFIIEHCIRMFKHYKQQRGKQKWFMK QNTEESTIGRKLSDHKLNNTPDSDWLQLKPLAGTDDSVVSEDRLNETELTDLEGQQESPP KNYLCIEEEKIIDHSHSDGLHTIHEHDLHAAAHNHHGENKTVLRKHNHQWHHKHSHHS HGPCHSGSDLKETGIANIAWMVIMGDGIHNFSDGLAIGAAFSAGLTGGISTSIAVFCHELP HELGDFAVLLKAGMTVKQAIVYNLLSAMMAYIGMLIGTAVGQYANNITLWIFAVTAGM FLYVALVDMLPEMLHGDGDNEEHGFCPVGQFILQNLGLLFGFAIMLVIALYEDKIVFDIQ F >XP_016805738.1 zinc transporter ZIP10 isoform X3 [Pan troglodytes] MKVHMHTKFCLICLLTFIFHHCNHCHEEHDHGPEALHRQHRGMTELEPSKFSKQAAENE KKYYIEKLFERYGENGRLSFFGLEKLLTNLGLGERKVVEINHEDLGHDHVSHLDILAVQE GKHFHSHNHQHSHNHLNSENQTVTSVSTKRNYKCDPEKETIEVSVKSDDKHMHDHNHR LRHHHRLHHHLDHNNTHHFHNDSITPSERGEPSNEPSTETNKTQEQSDVKLPKGKRKKK GRKSNENSEVITPGFPPNHDQGEQYEHNRVHKPDRVHNPGHSHVHLPERNGHDPGRGH QDLDPDNEGELRHTRKREAPHVKNNAIISLRKDLNEDDHHHECLNVTQLLKYYGHGAN SPISTDLFTYLCPALLYQIDSRLCIEHFDKLLVEDINKDKNLVPEDEANIGASAWICGIISIT VISLLSLLGVILVPIINQGCFKFLLTFLVALAVGTMSGDALLHLLPHSQGGHDHSHQHAH GHGHSHGHESNKFLEEYDAVLKGLVALGGIYLLFIIEHCIRMFKHYKQQRGKQKWFMK QNTEESTIGRKLSDHKLNNTPDSDWLQLKPLAGTDDSVVSEDRLNETELTDLEGQQESPP KNYLCIEEEKIIDHSHSDGLHTIHEHDLHAAAHNHHGENKTVLRKHNHQWHHKHSHHS HGPCHSGSDLKETGIANIAWMVIMGDGIHNFSDGLAIGAAFSAGLTGGISTSIAVFCHELP HELGDFAVLLKAGMTVKQAIVYNLLSAMMAYIGMLIGTAVGQYANNITLWIFAVTAGM FLYVALVDMLPEMLHGDGDNEEHGFCPVGQFILQNLGLLFGFAIMLVIALYEDKIVFDIQ F >NP_766241.2 zinc transporter ZIP10 precursor [Mus musculus] MKVHIHTKFCLICLLTFIFHHCNHCHEDHDHGPEELHRHHRGMTESESSKFSVQDAENEK KYYIEKLFDRYGENGRLSFFGLEKLLTNLGLGEIKVVEINHEDLGHDHVSHLDILAVQEG KHFHSHTHQHFHNHLNAENHTTTSVTSKRNHKCDPEKEAAELPIKADDKHLHDRNHRF HHRHRLHHHLDHNTTRHVHNDSVAHSEHGEPGHSPSPETNKTQEQSEVKSVKVRRKEK GKRKKENSEVNTPGFLPNHDHSEQYEHNRVHKLDRVHSPGHPHAHLPEHSGHELGHGH QELDPDNEGELRHTRKREAPHVRKSAIYSTPSHKDQSEDDRQHECLNVTQLLKHFGLGP NSPISPDLFTYLCPALLYQIDSRLCIEHFDKLLVEDLNKDKTLVPEDKTNIGASAWICGIISI TVISLLSLLGVILVPIINQGCFKFLLTFLVALAVGTMSGDALLHLLPHSQGGHDHSHQHTH GHGHSHGHESKEFLEEYDAVLKGLVALGGIYLLFIIEHCIRMFKHYKQQRGKQKWFMKQ 162 STEESTIGRKLSDHKLNSTPDADWLQLKPLAGTDDSVVSEDRLNETELTDLEAQQESPPK NYLGVEEEKIMDHSHSDGLHTIHEHEVHVTSHNHHDEDKAVLRKHSHQWHHRHAHHS HGPCHSGSDLKETGIANIAWMVIMGDGIHNFSDGLAIGAAFSAGLTGGISTSIAVFCHELP HELGDFAVLLKAGMTVKQAIVYNLLSAMMAYIGMLIGTAVGQYANNITLWIFAITAGM FLYVALVDMLPEMLHGDGDHEEHGFCPVGQFILQNLGLLFGFAIMLVIALYEDKIVFDIQ F >NP_001102266.2 zinc transporter ZIP10 precursor [Rattus norvegicus] MKVHIHTKFCLICLLTFIFHHCNHCHEDHDHGPEELHRHHRGMTESESSTFSVQDAENEK KYYIEKLFDRYGENGRLSFFGLEKLLTNLGLGEIKVVEINHEDLGHDHVSHLDILAVQEG KHFHSHNHQHFHNHLNAENHTATSITSKRNHKCDPEREAAEVPIKPEDKHPHDRNHRFH HRHRLHHHLDHNTTRRVHNDSIAHSEHGEPGHGPSTETNKTQEQSEVKSGKVRRKEKG KRKKENSEVNTPGFLPTHGHGEQYEHNRVHKLDRVHSPGHPHAHLPEHSGHELGHGHQ EFDPDNEGELRHTRKREAPHVKKSAIYSTPSHKDHNEDDRQHECLNVTQLLKHFGLGPS SPISPDLFTYLCPALLYQIDSRLCIEHFDKLLVEDLNKDKALVPEDKANIGASAWICGIISIT VISLLSLLGVILVPIINQGCFKFLLTFLVALAVGTMSGDALLHLLPHSQGGHDHSHQHAH GHGHSHGHESKKFLEEYDAVLKGLVALGGIYLLFIIEHCIRMFKHYKQQRGKQKWFMK QSTEESTIGRKLSDHKLNSTPDADWLQLKPLAGTDDSVVSEDRLNETELTDLEAQQESPP KNYLGVEEEKIMDHSHSDGLHTIHEHEVHVVSHNHRDEEDTAVLRKHGHQWHHRHAH HSHGPCHSGSDLKETGIANIAWMVIMGDGIHNFSDGLAIGAAFSAGLTGGISTSIAVFCHE LPHELGDFAVLLKAGMTVKQAIVYNLLSAMMAYIGMLIGTAVGQYANNITLWIFAITAG MFLYVALVDMLPEMLHGDGDHEEHGFCPVGQFILQNLGLLFGFALMLVIALYEDRIVFD IQF