ABSTRACT Title of Document: CYTOLOGICAL LOCALIZATION OF HEME IN CAENORHABDITIS ELEGANS USING MICROSCOPY Cornelia Elefson Keley, Masters degre, 2007 Directed By: Dr. Iqbal Hamza, Asistant Profesor, Department of Animal and Avian Sciences This study was designed to develop an in situ histochemical heme staining method for an intact animal using the fre-living nematode Caenorhabditis elegans. Although, heme is vital to many biological proceses and synthesized by most known fre-living organisms, C. elegans is a natural heme auxotroph. Using the substrate 3-3? diaminobenzidine and hydrogen peroxide, we used C. elegans to study the fate of heme. We found a direct correlation betwen heme in the growth medium and the organismal heme content. In addition, our studies confirmed that parents exposed to diferent heme levels contribute varying maternal heme to their progeny. Moreover, this methodology detected diferences in heme levels betwen wild-type and mutants in heme homeostasis. Finaly, we provide preliminary evidence that the technique can be applied to analyze heme-based structures at the electron microscopy level. Our studies described herein wil aid in the characterization of heme transport pathways in eukaryotes. CYTOLOGICAL LOCALIZATION OF HEME IN CAENORABDITIS ELGANS USING MICROSCOPY. By Cornelia Elefson Keley Thesis submited to the Faculty of the Graduate School of the University of Maryland, College Park, in partial fulfilment of the requirements for the degre of Master of Science 2007 Advisory Commite: Asistant Profesor Iqbal Hamza, Chair Profesor Ian Mather Profesor Tom Porter ? Copyright by Cornelia Elefson Keley 2007 i Dedication To my loving husband and precious daughter ii Acknowledgements Thank you to my advisor Iqbal, from whom I?ve learned such an incredible amount during my time here. To the other two members of my thesis commite, Ian and Tom. Thank you for your guidance. To my wonderfully patient lab mates from whom I learned so much: Anita, Abbhi, Caiyong, Melisa, Caitlin, Suji, Scott, Jason, and Sandy. Thank you especialy to Anita, Abbhi, Caitlin, and Scott who offered such valuable fedback on my literature review. Thank you al for putting up with me! Thank you to my family: my parents and sisters for al their encouragement and support. Thank you especialy to my mother whose numerous long trips from home to take care of her grandbaby were a tremendous help to Doug and myself. To al the graduate students of this department upon whose friendship, sense of humor, and advice I relied on throughout my time here. To al my family and friends who were so wonderfully supportive and never hesitated to lend a hand: Dad, Julia, Laura, Ann, Conoly, Melanie, Haynes, Jennifer, and Karin. To Jesie, for being the most wonderful and precious daughter any mother could ask for. And finaly to my husband Doug, if not for his loving support, encouragement, and patience this would not have been possible. iv Table of Contents Dedication.........................................................i Acknowledgements.................................................ii Table of Contents...................................................iv List of Tables......................................................vi List of Figures.....................................................vii List of Abbreviations...............................................vii Chapter 1: Literature Review...........................................1 Introduction.....................................................1 Heme biosynthesis................................................3 Regulation of heme biosynthesis.....................................7 Deficiencies in heme biosynthesis....................................9 Hemoproteins..................................................10 Heme as a regulator..............................................12 Heme degradation...............................................15 Heme transport.................................................16 Techniques used to localize heme...................................26 Chapter 2: Materials and Methods......................................29 Worm strains and growth medium...................................29 orm permeabilization and fixation for light microscopy.................29 Pre-staining solution to eliminate oxygen radicals.......................31 Heme staining for light microscopy..................................31 Heme staining analysis...........................................32 DAPI staining..................................................33 Worm fixation and permeabilization using microwave fixation.............33 Heme staining for electron microscopy...............................34 Worm embedding...............................................35 Thin sectioning.................................................36 Chapter 3 Characterization of heme localization in wild-type worms using light microscopy.......................................................38 Summary......................................................38 Rationale......................................................39 Results........................................................40 Addition of catalase to pre-incubation solution........................41 Addition of SOD to pre-incubation solution...........................44 Heme staining unsynchronized worms grown at 20 ?M and 100 ?M heme...45 Heme staining in synchronized L4 larvae at 20 ?M and 100 ?M heme......49 Synchronized L4 larvae grown at 1.5 ?M, 4 ?M, 20 ?M, and 40 ?M heme were stained for heme using deoxygenated conditions.......................49 Mixed population of worms grown at 1.5 ?M, 4 ?M, 20 ?M, and 40 ?M heme58 Standardization of necesary SOD concentration.......................60 Chapter 4: Application of new heme staining technique......................65 Summary......................................................65 v Rationale......................................................66 Results........................................................68 Worms mutated in heme homeostasis display diferential heme staining as compared to wild-type worms.....................................68 Progeny obtained from worms grown at diferent heme levels stain diferentialy for heme......................................................69 Preliminary results of heme staining C. elegans at the ultrastructure level....79 Discussion.....................................................87 REFERENCES.................................................95 vi List of Tables Table 1: Larval worms have significantly les non-specific staining than gravid worms.......................................................48 Table 2: Synchronized L4 larvae exhibit specific heme staining...............50 Table 3: Syncronized L4 larvae exhibit significant levels of specific heme staining when stained under deoxygenated conditions..........................55 Table 4: Diferences in specific heme staining betwen larval and gravid worms at 1.5 ?M, 4 ?M, 20 ?M, and 40 ?M heme.................................59 Table 5: Complementation groups of heme-resistant mutant worms.............70 Table 6: Percent of specific heme staining in wild-type worms and worms mutated in heme homeostasis...............................................71 Table 7: Heme resistant mutants reveal diferential heme staining..............72 Table 8: Maternal heme efect on growth rates of progeny obtained from parental worms grown at diferent heme concentrations.........................76 vii List of Figures Figure 1: Eukaryotic heme biosynthetic pathway............................4 Figure 2: Heme acquisition by gram-negative bacteria.......................18 Figure 3: Addition of CAT to eradicate peroxide radicals.....................42 Figure 4: Addition of SOD to eradicate oxygen radicals.....................46 Figure 5: Synchronized L4 larvae grown at 20 ?M and 100 ?M heme exhibit diferences in heme staining intensity................................51 Figure 6: Experimental apparatus for deoxygenation of worm samples...........53 Figure 7: Synchronized L4 larvae grown at 1.5 ?M, 4 ?M, 20 ?M, and 40 ?M heme and stained for heme under deoxygenated conditions....................56 Figure 8: Intestinal punctate staining....................................61 Figure 9: Heme absorption in intestinal cels..............................63 Figure 10: Heme staining wild-type, IQ731, IQ911, IQ828, IQ1068, and IQ938 strains grown at 4 ?M heme.......................................73 Figure 11: Progeny obtained from worms grown at diferent heme levels reveal a maternal heme efect............................................77 Figure 12: Comparison of microwave fixation to chemical fixation.............81 Figure 13: Wild-type worms grown at 80 ?M heme, histochemicaly stained for heme, and examined at the ultrastructural level.........................85 vii List of Abbreviations ALA ??aminolevulinic acid ALAS ??aminolevulinic acid synthase IR Iron response regulator Sre1p Sterol regulatory element binding protein NPAS2 Neuronal Pas domain protein 2 BMAL1 Brain and muscle arnt-like protein-1 IRE Iron responsive element IRP Iron responsive protein CDK Cyclin-dependent kinase NGF Nerve growth factor DGCR8 DiGeorge critical region-8 HO Heme oxygenase Hb Hemoglobin RBCs Red blood cels OGC 2-Oxoglutarate carier FLVCR Feline leukemia virus subgroup C celular receptor PdTCP Paladium,meso-tetra[4-carboxyphenyl]porphyrin PdTCP Paladium, meso-tetra(4-aminophenyl)porphyrin PdMP Paladium?mesoporphyrin IX EM Electron microscope N2 C. elegans wild-type strain IQ1068d Mutant worm strain IQ1068 dumpy DAB 3-3? diaminobenzidine mCeHR Modified Caenorhabditis elegans Habitatation and Reproduction D Day(s) H Hour(s) Min Minutes PBS Phosphate buffered saline SOD Superoxide dismutase CAT Catalase 1 Chapter 1: Literature Review Introduction The porphyrin, heme is one of the most biologicaly diversified molecules in nature. Heme functions as a prosthetic group in hundreds of proteins including the globins, respiratory cytochromes, cytochrome P450s, and thyroperoxidases, al proteins vital to proceses such as oxygen transport, xenobiotic detoxification, oxidative metabolism, and thyroid hormone synthesis (1). In addition to functioning as a prosthetic group for a wide variety of proteins, heme also directly regulates biological proceses such as celular diferentiation, cel cycle progresion, and gene regulation (2,3). Most eukaryotes and prokaryotes synthesize heme in a highly conserved eight-step biosynthetic pathway. Regulation of heme biosynthesis has been extensively studied in many organisms including bacteria, yeast, and mamals (4-6). In humans, a deficiency in any of the enzymes of the heme biosynthetic pathway can lead to diseases known as the porphyrias. There is a considerable amount of information known about eukaryotic heme biosynthesis and the wide-range of heme?s functions in eukaryotes, however, very litle is known about how this molecule is transported into cels, within cels, and into celular organeles. While there have been studies that traced heme absorption in mamalian intestinal cels (7,8), only recently has research begun to identify proteins involved in eukaryotic heme uptake and transport (9-12). Once heme is synthesized in the mitochondria, some of it is used for incorporation into mitochondrial 2 hemoproteins. The rest of the newly synthesized heme must be transported out of the mitochondria for incorporation into hemoproteins located in the cytosol, nucleus, endoplasmic reticulum, and other organeles in order to participate in various celular proceses. Heme is hydrophobic and cytotoxic and can cause extensive celular damage due to its inherent peroxidase activity. Thus, a prima facie argument can be made that heme likely does not exist fre in cels. Understanding the fundamental biological proces of eukaryotic heme traficking has wide-ranging implications for solving presing human, livestock, and crop health isues. From a human nutritional standpoint, dietary heme is the most bioavailable form of iron, but the heme absorption pathway(s) into intestinal cels is unknown. From a disease standpoint, helminthic infections are a tremendous burden to humans and livestock. Recent research indicates that several species of infectious nematodes are heme auxotrophs, i.e. they cannot synthesize heme but acquire it from diet (13). Discovery of heme transport pathway genes unique to these helminths could lead to the development of new pharmaceuticals to control helminthic infections. Several techniques have ben developed to cytologicaly localize heme and characterize its absorption into and transport within cels. Many methods involve the use of fluorescent dyes (14), fluorescent heme analogs (15), and radioisotopes (16), but one wel-established method can be used to directly stain heme. This staining method involves the use of the compound 3-3? diaminobenzidine (DAB) and exploits the peroxidase activity of heme to generate a localized reaction product (17). Unpublished results have established that the non-parasitic nematode C. elegans is a natural heme auxotroph and thus provides an ideal model organism for studying heme 3 uptake (13). Establishment and examination of staining whole animals, such as C. elegans, for heme wil provide a new approach to expand current knowledge of organismal heme homeostasis. Heme biosynthesis Heme biosynthesis ocurs in a complex eight-step pathway conserved betwen eukaryotes and prokaryotes and can be broken down into several key proceses (18). Briefly, a pyrrole ring is formed by two molecules of ?- aminolevulinic acid (ALA), followed by their asembly into a tetrapyrrole. Protoporphyrinogen IX is then formed by modification of the tetrapyrrole side chain and is subsequently oxidized to protoporphyrin IX. Iron is inserted into protoporphyrin IX in the final step to form protoheme, or heme b, the most common form of heme (18). Depending upon the hemoprotein for which heme is needed, heme b can then be modified to heme a or heme c. The first step in eukaryotic heme biosynthesis occurs in the mitochondria and results in the formation of ALA, the universal porphyrin precursor, by a condensation reaction betwen succinyl CoA and glycine catalyzed by ALA synthase (ALAS), as sen in Figure 1. ALA is then transported out of the mitochondria into the cytoplasm. In the second step, two molecules of ALA form porphobilinogen in a condensation reaction catalyzed by aminolevulinic acid dehydratase, also known as porphobilinogen synthase. Four molecules of porphobilinogen are then used to form the unstable polymer hydroxymethylbilane in the third reaction catalyzed by porphobilinogen deaminase. Uroporphyrinogen II synthase then converts 4 Figure 1: Eukaryotic heme biosynthetic pathway. The eight-step eukaryotic heme biosynthetic pathway begins with the synthesis of ALA in the mitochondria. ALA is then transported from the mitochondria into the cytosol, where the subsequent four steps occur. The heme precursor coproporphyrinogen II is transported back into the mitochondria for the final thre steps culminating in the formation of protoheme (heme b). Mitochondrion Cytosol Glycine + Succinyl CoA ALA synthase ALA 1 ALA ALA dehydratase Porphobilinogen Porphobilinogen deaminase Hydroxymethylbilane 2 3 4 5 Uroporphyrinogen III Uroporphyrinogen II synthase Uroporphyrinogen decarboxylase Coproporphyrinogen III Coproporphyrinogen oxidase 6 Protoporphyrinogen IX Protoporphyrinogen oxidase Protoporphyrin IX 7 ferochelatase Protoheme 8 5 6 hydroxymethylbilane to uroporphyrinogen II during step four. Step five involves the conversion of uroporphyrinogen II to coproporphyrinogen II catalyzed by uroporphyrinogen decarboxylase. Coproporphyrinogen II is then transported back into the mitochondria where coproporphyrinogen oxidase, an enzyme located in the mitochondrial inner membrane, catalyzes the formation of protoporphyrinogen IX during step six. Next, protoporphyrin IX is formed by protoporphyrinogen oxidase. The final step occurs when ferous iron is inserted into the protoporphyrin IX cyclic macromolecule to form heme b. This final reaction is catalyzed by ferochelatase. Further modifications of heme b at C-2 and C-8 are required to form heme a, the form of heme found in cytochrome oxidases (19). Synthesis of heme a from heme b involves two enzymatic reactions. Heme O synthase catalyzes the conversion of the vinyl group to a 17-hydroxyethylfarnesyl to generate heme o. Heme o is then converted to heme a by heme A synthase (HAS) (20). Heme c is found in cytochromes c1 and c and difers from heme b in that heme c covalently binds to its proteins through the two vinyl side chains (21). While this biosynthetic pathway is highly conserved betwen eukaryotes and prokaryotes, there are a few diferences. One obvious diference is that, in eukaryotes, the biosynthetic pathway is partialy compartmentalized betwen the mitochondria and the cytosol, as demonstrated in Figure 1. Another diference is that some prokaryotes can synthesize ALA by an alternative pathway. These organisms synthesize ALA via the C-5 pathway, a pathway that converts glutamic acid to ALA acid glutamyl-tRNA and glutamate semialdehyde as intermediates (22). No known prokaryote has the capability to synthesize ALA using both the C-5 pathway and the 7 ALAS pathway. One other notable diference betwen eukaryotes and prokaryotes is that eukaryotic heme demand requires heme to be both exogenously acquired as wel as endogenously synthesized, while prokaryotic heme demand is usualy met by heme biosynthesis (23). Regulation of heme biosynthesis Regulation of heme biosynthesis has been studied in many diferent model systems and varies from organism to organism. In some organisms, heme biosynthesis is connected to iron availability (4,18), while, in others, regulation can depend on oxygen levels (5). In the yeast Schizosaccharomyces pombe, Sterol Regulatory element binding protein (Sre1p) up-regulates six heme biosynthetic pathway enzymes during anaerobic conditions (5). It is thought that Sre1p maintains heme levels under oxygen-limiting conditions. In the bacterium Bradyrhizobium japonicum, heme biosynthesis is coordinated to iron availability (4). Iron is inserted in the protoporphyrin IX ring by ferochelatase during the final step of the heme biosynthetic pathway. During iron deficiency, bacterial cels with a mutation in the iron response regulator (IR) gene are unable to arest the heme biosynthetic pathway and acumulate toxic levels of protoporphyrin IX. This protein coordinates heme biosynthesis to iron availability. Additional studies showed that IR inhibits the heme biosynthetic pathway by inhibiting hemB, the gene encoding ALA dehydratase in B. japonicum. Under iron limiting conditions, IR inhibition of hemB ultimately prevents toxic acumulation of protoporphyrin IX in wild-type organisms (4). 8 In mamals, it is the first step of heme biosynthesis, catalyzed by the enzyme ALAS, that is highly regulated. Mamals difer from other organisms in that they require two diferent forms of ALAS. While heme biosynthesis occurs in every cel, the majority of heme biosynthesis occurs in the bone marow for incorporation into hemoglobin. This increased demand for heme biosynthesis in the bone marow requires a separate form of ALAS, ALAS2. ALAS2 is expresed only in erythroid cels, in the bone marow (24), while ALAS1 encodes the constitutively expresed form of ALAS. Regulation of heme biosynthesis in erythroid cels via ALAS2 is thus separate from regulation of constitutive biosynthesis at the level of ALAS1 (18). In mamals, the main transcriptional regulator of ALAS1 is neuronal Pas domain protein 2 (NPAS2), a mamalian transcription factor known to be part of the core circadian clock mechanism (6). NPAS2 is a PAS domain protein, a family of proteins that can detect environmental signals such as oxygen and light, as wel as smal aromatic molecules, including heme (25). NPAS2 has two PAS domains, each of which bind heme, and the ability of NPAS2 to bind DNA is dependent on heme availability. When NPAS2 binds heme, it forms a heterodimer with BMAL1, and this transcription complex binds the ALAS1 promoter. Expresion of ALAS1 is then upregulated, thus inducing heme biosynthesis (6). Exces heme in the cel induces heme oxygenase, the enzyme responsible for degrading heme into carbon monoxide, biliverdin, and iron. In the presence of carbon monoxide, the NPAS2 ?BMAL1 transcription complex cannot sufficiently bind DNA, thus preventing expresion of ALAS1 (25). In erythroid cels, ALAS2 is transcriptionaly regulated by histone 9 acetyltransferase p300 (26). This transcriptional regulator is recruited by the erythroid specific transcription factor, GATA1. The interaction of these two proteins activates the ALAS2 promoter (27). Translational regulation of ALAS2 in murine erythroleukemia cels is regulated in an iron?dependent manner by the Iron Responsive Element/ Iron Responsive Protein (IRE/IRP) system (28). When iron is low, IRP binds to the IRE of the 5? untranslated region of the ALAS2 mRNA transcript, inhibits translation and prevents heme synthesis in these erythroid cels. While IRP exists in two forms, IRP1 and IRP2, only IRP2 translationaly regulates ALAS2 (29). IRP2 knockout mice overexpres ALAS2 and have elevated levels of fre protoporphyrin IX in their red blood cels, which is possibly responsible for observed phenotypes that include photosensitivity, skin and eye lesions (29). Deficiencies in heme biosynthesis Any deficiency in ALAS2 or the other seven enzymes in this pathway can cause the porphyrias. Porphyrias are a family of diseases that cause neuropsychiatric symptoms including anxiety, insomnia, confusion, halucinations, agitation and paranoia (30). Porphyria atacks could occur because a) heme deficiency leads to deficiency of hemoproteins critical to celular function, or b) enzyme defects cause toxic levels of pathway intermediates to acumulate, resulting in a neurotoxic efect (30). An additional explanation is that heme deficiency results in decreased production of microRNAs, a clas of RNAs vital to many diverse biological proceses. Recently, microRNA procesing was shown to involve heme (3). Heme deficiency could have a global efect by impacting a number of biological proceses. 10 Hemoproteins Once heme is endogenously synthesized or exogenously acquired, it must then be transported to the cytoplasm or celular organeles for insertion into any of the hemoproteins vital to celular function. The cytochrome P450 enzymes are hemoproteins whose primary role is xenobiotic detoxification. Xenobiotics are foreign chemicals found in an organism that must be safely metabolized and excreted. Food additives, industrial chemicals, pesticides, plant toxins and pharmaceutical agents are common xenobiotics. Deamination, desulfuration, hydroxylation, and oxidation reactions catalyzed by cytochrome P450s alow cels to metabolize xenobiotics into a form that can be excreted (31). In yeast and humans, cytochrome P450s are required for sterol biosynthesis. It was discovered recently that the eukaryotic hemoprotein Dap1 binds several endoplasmic reticulum cytochrome P450 proteins in both humans and yeast and that Dap1 is esential for sterol biosynthesis (32). Another family of hemoproteins is the heme-based sensor family. These proteins are key regulators of adaptive responses to changing levels of oxygen, carbon monoxide, and nitric oxide. Proteins are clasified as heme-based sensors because one protein domain is controlled by a heme-binding domain within the same protein (25). There have been more than fifty heme-based sensors studied in bacteria and mamalian cels (26). One example of a mamalian heme-based sensor is soluble guanylyl cyclase. This cytosolic protein synthesizes cyclic guanosine monophosphate (cGMP) from guanosine triphosphate (GTP) when it can detect 11 nitrous oxide and carbon monoxide. cGMP then acts on downstream efectors that control smooth muscle tone and neurotransmision among other physiological proceses (26). Heme is vital to celular respiration because several key hemoproteins are involved in energy production (1). These hemoproteins are the respiratory cytochromes, electron-transfering proteins located in the mitochondrial inner membrane. The iron atom at the center of the heme ring alternates betwen a reduced ferous state and an oxidized feric state during electron transport reactions. Two of these mitochondrial protein complexes are cytochrome reductase complex (complex II) and cytochrome c oxidase (complex IV). The cytochrome reductase complex contains thre heme prosthetic groups, while complex IV, or cytochrome c oxidase, has two heme prosthetic groups (1). Hemoproteins are also involved in thyroid hormone synthesis (33). Thyroperoxidase is a membrane-bound hemoprotein that catalyzes the production of the thyroid hormones, thyroxine (T 4 ) and 3,3? 5-triodothyronine (T 3 ). The traficking of thyroperoxidase to the cel surface from the endoplasmic reticulum and subsequent enzymatic activity is dependent on insertion of heme into apo-thyroperoxidase. Heme must therefore be transported into the ER for incorporation into thyroperoxidase. Inhibition of heme biosynthesis significantly reduces thyroperoxidase traficking to the cel surface and its subsequent enzymatic activity (33). 12 Heme as a regulator Heme also plays a regulatory role in many biological proceses including cel growth, cel cycle progresion, celular diferentiation, gene regulation, and microRNA procesing. This regulatory role can be cel-type specific. While heme promotes cel growth and cel cycle progresion in HeLa cancer cels, it promotes celular diferentiation in PC12 pro-neuronal cels and K562 erythroid cels (2). In HeLa cels, heme directly controls the cel cycle by acting on two key cel cycle regulators, cyclin-dependent kinases (CDKs) and p53 (34). CDK proteins promote cel cycle progresion, while p53 suppreses cel growth by activating cel cycle suppresor proteins and inducing BAX and Fas, two proapoptotic genes. When DNA in a cel becomes damaged, the functional protein product of p53 induces the transcription of p31, whose protein product blocks the activity of CDK complexes. Inhibition of CDK complexes prevents the cel from progresing to the S-phase of the cel cycle, preventing damaged DNA from replicating. When heme biosynthesis was inhibited in HeLa cels, they arested in the S phase of the cel cycle and this cel cycle arest was acompanied by morphological changes asociated with senescence. Addition of heme reversed these efects. Inhibiting heme synthesis increased p53 levels four to five fold, an efect that was reversed when heme was added back to the medium, suggesting that heme has a direct efect on p53. Conversely, inhibiting heme biosynthesis decreased the celular levels of thre CDK proteins: Cdk4, Cdc2, and cyclin D4. Adding heme to these cels again reversed these efects (34). In PC12 cels and red blood cels, heme plays a role in celular diferentiation. 13 Heme is critical for neuronal diferentiation via the Ras-ERK 1 and 2 signaling pathways (2). Heme deficiency inactivates the Ras-ERK signaling pathways that are induced by nerve growth factor (NGF), a protein secreted by neurons that plays a crucial role in the diferentiation of neural crest-derived sensory and sympathetic neurons. Heme deficiency caused NGF-diferentiated PC12 cels to undergo caspase activation and subsequent apoptosis. Experiments also showed that undiferentiated PC12 cels were les vulnerable to heme deficiency. This disparity in heme deficiency vulnerability betwen diferentiated and undiferentiated cels suggested a role for heme in NGF signaling (2). Additionaly, male mouse embryos hemizygous for ALAS2 die at embryonic day 11.5 due to severe anemia. The primitive erythroid cels did not mature, suggesting an additional role for heme in erythroid celular diferentiation (24). In addition to a direct role in controlling the cel cycle and celular diferentiation, heme also has an indirect role in regulating developmental timing proceses, celular diferentiation, and apoptosis through its involvement in microRNA production (3). MicroRNAs are short RNAs, generaly twenty-two nucleotides long, that play a significant role in regulating protein-encoding genes. Ten to thirty percent of al protein-encoding genes are regulated by microRNAs. Recently, it was shown that heme plays a role in the activation of a protein critical to microRNA procesing. This protein, DiGeorge critical region-8 (DGCR8), requires heme for dimerization and subsequent activation. After DGCR8 dimerizes, it binds to primary microRNAs, and the resulting trimer triggers cleavage of the primary microRNA by DROSHA to produce mature microRNAs (3). This particular 14 regulatory role for heme could have severe ramifications for cels during heme deficiency. In this case, heme deficiency would result in a reduction of mature microRNAs, an efect that could seriously impact multiple celular proceses. Recent work suggests that there is reciprocal regulation betwen heme biosynthesis and the circadian clock (6). Circadian clocks play a crucial role in regulating physiology and behavior in organisms ranging from plants to animals (35). Research indicates that heme control of the circadian clock complex is transcriptionaly regulated (6). In Drosophila, the dperiod gene is an intrinsic part of the animal?s circadian clockwork. Two dperiod homologs, mPER1 and mPER2 regulate the mamalian circadian clock. In mice, heme was shown to decrease expresion of mPer2 but increase expresion of mPer1, an efect most pronounced when the mice were subjected to complete darknes. These results suggested involvement in the circadian clock. Further studies showed that the expresion paterns of mPer1 and mPer2 were diferentialy heme-regulated by NPAS2 and mPER2, two circadian clock proteins that bind heme. When heme binds NPAS2, NPAS2 forms a transcription complex with BMAL1, a complex positively regulated by mPER2 in vivo. This complex activates the ALAS1 promoter, thus inducing heme biosynthesis (6). In yeast, heme regulates gene expresion through the heme-responsive transcriptional regulator, Hap1p (5). In S. cerevisiae, Hap1p regulates the expresion of ROX1, a gene that regulates expresion of roughly one-third of al anaerobicaly expresed genes. Under aerobic conditions, ROX1 transcription is upregulated by Hap1p, and genes necesary for anaerobic conditions are represed. Extracelular 15 oxygen levels determine heme synthesis in yeast, and, under long-term hypoxic conditions, oxygen-dependent heme synthesis decreases. This is followed by a decrease in HAP1p activity and subsequent down-regulation of ROX1, which alows the expresion of genes needed for hypoxic conditions (5). Hap1p is also involved in steroid biosynthesis in yeast in two ways. One is by regulating HMG1, one of the genes in the sterol biosynthetic pathway, and the other is by playing a role regulating gene targets of the two transcription factors Upc2p and Ecm22p (36). Upc2p and Ecm22p activate the ERG genes, which are responsible for sterol biosynthesis in yeast. Under hypoxic conditions, Hap1p plays a role in regulating gene targets of Upc2p and Ecm22p, but precisely how it does this is not entirely clear. Basal expresion of another sterol biosynthetic gene, ERG2, requires Hap1p, regardles of whether the expresion was activated by Ecm22p or Upc2p. It is not clear whether sterols or heme are responsible for induction of these transcription factors under hypoxic conditions. However, hypoxic induction of ERG genes responds to heme levels rather than sterol levels, a response that does not involve either Upc2p or Ecm22p (36). Heme degradation While heme biosynthesis is absolutely esential to celular function, heme degradation is also an important aspect of celular metabolism. The main purpose of heme degradation is the reutilization of iron from hemoglobin (31). Heme is degraded by heme oxygenase (HO) to ferous iron, carbon monoxide and biliverdin. There are two isoforms of heme oxygenase, HO-1 and HO-2, each encoded by a 16 separate gene. HO-2 is the constitutive form of heme oxygenase, while HO-1 is highly inducible (2). HO-1 is regulated by Bach1, a heme-regulated mamalian transcriptional represor (37). In human hepatocytes, silencing Bach1 upregulates HO-1 gene expresion. Bach1 downregulates HO-1 by forming heterodimers with Maf proteins and binding to the heme responsive elements (HeRes) in the 5? untranslated region of the HO-1 promoter. Heme binds to Bach1, causing a conformational change that markedly decreases the ability of Bach1 to bind HeRe elements. When the Bach1 complex no longer binds these binding sites on the promoter, the sites can then be bound by the HO-1 transcriptional activators, Maf-Maf and Nrf2-Maf (37). Heme transport Heme transport has been studied in a wide variety of organisms including bacteria, yeast, invertebrates, and mamals. However, this field of research has ben complicated because many model organisms are heme prototrophs, i.e. they synthesize heme. Distinguishing betwen heme synthesized intracelularly and heme obtained from exogenous sources, therefore, is dificult without labeling one source of heme with fluorescent dyes, fluorescent molecules, or radioisotopes. While prokaryotic heme biosynthesis is sufficient for prokaryotic heme demand (23), many bacteria obtain heme from exogenous sources to utilize as a source of iron (38). Many proteins directly responsible for heme uptake and transport in bacteria have been identified and characterized (38). In contrast to the field of bacterial heme transport, eukaryotic heme transport 17 proteins have been identified only recently. Two genes with homologs in other pathogenic yeast species are required for heme uptake in the yeast Candida albicans (39). In 2005, it was discovered that the model organism, C. elegans requires heme but does not synthesize it. While C. elegans is non-pathogenic, there are pathogenic nematodes that also require heme but are unable to synthesize it (13). These findings suggest that heme uptake could be an Achiles hel for some nematodes harmful to human, crop, and livestock health (13). The pathways for how heme is transported into cels from external sources, transported betwen cels, and transported within cels, remains to be identified. Heme uptake has been wel characterized in many bacterial species and the general mechanisms by which this occurs in gram-negative bacteria is demonstrated in Figure 2 (38). Gram-positive bacteria contain specific binding proteins anchored to the inner membrane that recognize heme sources. Once these heme sources are identified and bound, they are transported by ABC permeases. The first involves direct contact betwen heme or heme binding proteins and an outer-membrane receptor while the second system involves the release of proteins known as hemophores into the extracelular media by the bacteria. The hemophores bind heme or heme-binding proteins such as hemoglobin and hemopexin, and deliver them to specific receptors located on the bacteria?s outer membrane. The hemophore system has thus far been found only in gram-negative bacteria (40). Two of these hemophore systems are the HxuA and HasA systems (38). The HxuA system is found in Haemophilus influenza and is the only bacterial system known to utilize heme bound to hemopexin (38). H. influenza does not synthesize 18 Figure 2: Heme acquisition by gram-negative bacteria. Bacteria secrete hemolysin to degrade red blood cels (RBC?s), releasing hemoglobin and heme with heme being bound by albumin or hemopexin. There are then thre diferent ways in which the bacteria obtain the heme. 1 Heme and hemoglobin can bind directly to a TonB-dependent outer membrane receptor. 2 Bacteria release proteins caled hemophores that capture hemoglobin and hemopexin. The hemophores then bind to specific TonB dependent outer membrane receptors. 3 Bacterial proteases, either membrane-bound or secreted, degrade hemoglobin, hemopexin, and albumin. This proces releases heme that subsequently binds to specific TonB ?dependent receptors and is released into the periplasmic space. Heme is then transported through the cytoplasmic membrane by a proces that is not wel described. Once heme reaches the cytoplasm it is degraded by heme oxygenase, releasing iron and protoporphyrin IX (adapted from reference (38)). hemolysin hemophore Protease (secreted) Hb hemeRBC 1. 2. 3. Protease (membrane- bound) Heme oxygenase IRONPPIX PP OM CM permease ATPase 19 OM = outer membrane P = periplasmic space CM = cytoplasmic membrane 20 heme, and contains the HxuABC gene cluster that alows it to eficiently utilize hemopexin, a heme-binding blood plasma protein. HxuA is secreted, binds hemopexin and this complex binds to HxuA-hemopexin specific receptors. The HasA system is found in at least five bacterial species and functions by capturing fre heme or heme from hemoglobin and presenting it to specific outer-membrane receptors (40). Animals combat parasitic infections by limiting the amount of fre iron in their blood. Because of this limited availability of iron, many pathogenic microorganisms have evolved heme acquisition pathways that alow them to use heme from their host as a source of iron, and at least one species of pathogenic bacteria prefers heme iron over iron bound to transferin (41). Also, it has been shown that targeting heme uptake pathways of pathogenic bacteria can afect their virulency (41,42). Mutating genes involved in the heme uptake system of the bacteria responsible for whooping cough in humans, Bordatela pertussis, reduced its infective strength in the mouse because the organism?s iron uptake systems were unable to compensate for the inability to acquire heme (42). Mutating the heme uptake pathway in the gram-positive bacterium, Staphylococcus aureus, reduced its infective strength in the worm model, C. elegans (41). While proteins involved in bacterial heme uptake and utilization have been studied extensively, only recently have proteins with similar functions been discovered in eukaryotes. Two recently discovered genes, RBT5 and RBT51, are responsible for heme uptake in C. albicans (9). While these two proteins have a seventy percent identity to one another, RBT5 plays the dominant heme acquisition role in C. albicans. However, RBT5 does not 21 confer to another yeast species, S. cerevisiae, the ability to utilize hemoglobin as an exogenous iron source (9). It was shown that Rbt5p and Rbt51p are heme receptors induced by iron starvation. While these genes alow C. albicans to utilize heme as an iron source, the underlying molecular mechanism is not known. One proposed mechanism is that these proteins facilitate the difusion of heme through the plasma membrane. Another hypothesis is that Rbt5p could alow the heme to be taken up by a not-yet- discovered heme transporter. Alternatively, Rbt5/Rbt51 could be internalized from the plasma membrane with the heme where the heme is then released into the cytoplasm (9,43). In addition to RBT5 and RBT51, C. albicans contains another gene esential to its ability to utilize hemoglobin as an iron source. The gene, CaHMX1, is a heme oxygenase that is twenty-five percent homologous to human heme oxygenase. While CaHMX1 is the first gene known in C. albicans to be regulated by heme (44) and mamalian hemoglobin (45), its exact enzymatic activities are not yet clear. While C. albicans can become pathogenic and use heme as an iron source, mutating genes responsible for heme uptake did not reduce its virulency in the mouse model. Two additional pathogenic yeast species, however, eficiently utilized heme as an iron source and the proteins responsible for this are homologous to Rbt51p, indicating a possible vulnerability of these pathogens to heme deficiency (9). Worldwide, pathogenic nematodes cause significant problems for humans, crops, and livestock. One-third of Earth?s human population is infected with soil- transmited helminths and the majority of those infected live in the developing world (46) Large-scale drug treatment programs have been designed and implemented to 22 combat these helminthic infections in humans, increasing the likelihood of drug resistance. Although anthelmintic drug resistance in humans has not yet become a problem, wide-spread use of anti-helminths in the animal husbandry industry has led to rampant drug resistance (47). Recent studies suggest that a potential Achiles hel and novel pharmaceutical target for some of these parasites is heme uptake (13). When it was discovered that the non-parasitic model organism C. elegans is a natural heme auxotroph, further studies revealed that five other phylogeneticaly diverse species of parasitic nematodes also depend on heme acquisition from exogenous sources (13). Another recently published report suggests that host dietary iron supply afects the ability of the hookworm Ancylostoma ceylanicum to infect its host (48). Hamsters infected with A. ceylanicum and fed a diet with severe iron restriction, had a significant reduction in the parasite load compared to animals that were fed a diet with moderate levels of iron. These results suggest that hookworms rely on the host iron to maintain growth and development. Host iron deficiency may lead to impaired host heme production, afecting the ability of hookworms to sustain infection. This discovery has potential ramifications for how humans are treated for hookworm infection. Traditionaly, patients infected with hookworms are at risk for becoming anemic and are given iron supplements as part of their treatment. Results from this study, however, suggest that this treatment could actualy be alowing the parasite to thrive in its host (48). Correspondingly, another nematode that may lack heme biosynthetic enzymes is Brugia malayi, a filarial nematode that infects one hundred and fifty milion people 23 worldwide (49). Genomic studies of two other filarial nematodes suggest the possibility that Brugia malayi requires heme for reproduction and development via hemoproteins that catalyze hormone biosynthesis. It appears that B. malayi may be another natural heme auxotroph and it likely obtains heme either from its host or possibly from Wolbachia, an endosymbiotic bacterium. Wolbachia contains al but one of the heme biosynthetic enzymes and is most likely responsible for supplying heme to B. malayi. In human trial studies, B. malayi infected patients have been succesfully treated using antibiotics that target Wolbachia. Targeting the heme uptake pathway in Brugia malayi is another potential pharmaceutical target as it appears that the survival and growth of this organism requires heme (49). The field of mamalian heme transport has only recently begun to identify genes involved in this proces. Two mamalian mitochondrial porphyrin transporters have been identified, one of which also appears to transport heme (10,12). Additional advances in mamalian heme transport have revealed two genes responsible for heme export in hematopoietic stem cels (11,50). One of the mamalian mitochondrial porphyrin transporters discovered is ATP-binding casete transporter ABCB6, which locates to the outer mitochondrial membrane and is required for mitochondrial porphyrin uptake (12). Experiments conducted in two mouse models demonstrated that the expresion of this protein directly responded to intracelular amounts of heme. Mice treated to enhance splenic erythropoiesis showed an increase in Abcb6 mRNA. Abcg2, also known as the breast cancer resistance gene (Bcrp1), is a member of the ATP-binding casete family of drug transporters that transports drugs and toxins from cels (51). In Abcg2 knockout mice 24 that have higher levels of protoporphyrin IX, mRNA levels of Abcb6 were significantly higher in the bone marow, liver, kidney, and testis (12). Heme biosynthesis, erythroid diferentiation, and intracelular levels of protoporphyrins were shown to directly influence levels of murine ABCB6 in vivo. In K562, SAOS-2, and MEL cels, overexpresion of Abcb6 had a six to thirten fold increase in acumulation of protoporphryn IX, an increase blocked by the heme synthesis inhibitor, succinyl acetone, suggesting that increasing Abcb6 expresion increased heme biosynthesis. ABCB6 asociates with the outer mitochondrial membrane and is thought to transfer coproporphyrinogen II to the enzyme coproporphyrinogen oxidase which resides on the inner mitochondrial membrane to facilitate heme biosynthesis (12). Another mitochondrial porphyrin transporter recently characterized is 2- oxoglutarate carier (OGC), an inner mitochondrial membrane protein known to facilitate uptake of 2-oxoglutarate into the mitochondria (10). The studies involved the use of two fluorescent porphyrin derivatives PdTCP and PdTAP. PdTCP acumulated in the mitochondria and co-localized with the mitochondrial marker, Mito Tracker. OGC was identified as the protein bound to PdTCP by using latex bead technique and mas spectrometry. It was then shown that 2-oxoglutarate uptake into the mitochondria was inhibited by porphyrin derivatives. Uptake of PdTCP into the mitochondria was prevented by 2-oxoglutarate, while hemin and two heme precursors, protoporphyrin IX and coproporphyrin II, inhibited 2-oxoglutarate uptake into the mitochondria. This carier protein is conserved in yeast and humans and may play a role in the acumulation of heme and heme precursors in the mitochondria 25 (10). In spite of these promising results, there are a few caveats in this study. One is the primary role of the synthetic fluorescent derivative PdTCP, a compound whose physiological properties are not clear. Another question to addres is why heme would be transported into the mitochondria. And finaly, in each of these papers (10,12), the authors used the commercialy available oxidized planar conjugated macrocycle, coproporphyrin II, rather than the physiologicaly relevant substrate coproporphyrinogen II, a reduced nonplanar porphyrin (52). Until recently, no proteins that specificaly transported heme had been identified. Within the last two years, two mamalian heme transporters were identified: the feline leukemia virus receptor (FLCVR) and the ABC transporter (ABCG2) (11,50). Both of these apical membrane proteins were shown to facilitate the export of heme from hematopoietic stem cels, preventing toxic levels of heme from acumulating. Preventing expresion of FLVCR by siRNA causes a decrease in the number of colony-forming unit erythrocyte cels, induces apoptosis, and subsequently prevents erythroid diferentiation. Expresion of FLVCR in CaCO-2 cels reduces celular heme concentration during intestinal cel diferentiation (31). While the role of FLVCR in intestinal and liver tisues remains unclear, ABCG2 localizes on the apical membrane of duodenal enterocytes and in other regions of the gastrointestinal tract. The definitive function of these proteins as heme exporters in the intestine remains to be explored. Currently, the protein or proteins responsible for dietary heme absorption into intestinal cels are unknown. While it was recently reported that the long-sought heme carier protein had been identified and subsequently named HCP1, a group 26 several months later reported that HCP1 is a proton?coupled folate transporter and that it could not be more than a poor heme transporter (53). Therefore, the search for a mamalian gene or genes that facilitate intestinal heme uptake continues. Techniques used to localize heme There have been many techniques developed to localize heme within cels, many of which involve fluorescent (14,15), histochemical (7,8), and radiolabeling approaches (16). Recently, heme traficking was observed in live cels for the first time in the catle tick, Boophilus microplus (15). This parasitic heme auxotroph relies on the blood of its host for nutrients and can ingest quantities of blood up to one hundred times its own body mas. During the ensuing breakdown of hemoglobin, potentialy toxic levels of heme are generated and sequestered into a recently described organele caled the hemosome (54). To observe heme traficking within the cel, researchers used the fluorescent heme analog Paladium?mesoporphyrin IX (Pd-mP) (15). Pd-mP fluoresces red when bound to proteins and gren when fre. This enabled the researchers to determine when and where the heme analog was bound to proteins in live digest cels of the B. microplus. Pulse-chase experiments revealed the heme analog was bound to protein while being traficked from digestive vesicles through the cytoplasm, and then transported into the hemosome. When the analog was present in the cytoplasm, it fluoresced red and this fluorescence intensified at the edge of the hemosome (15). Histochemical staining methods have also previously been established to pinpoint heme localization in mamalian tisue using the chromogenic substrate, 3- 27 3?diaminobenzidine (DAB) (7). This technique exploits the oxidative property of the ferous iron (FeI) within the tetrapyrrole ring. The iron in the center of heme reacts with hydrogen peroxide to generate hydroxyl radicals as shown in the following equations: Fe 2+ H 2 O 2 -> Fe 3+ . OH + OH- Fe 3+ H 2 O 2 -> Fe 2+ . OH + H + Hydrogen peroxide oxidizes FeI of heme to feric iron (FeII), a hydroxyl radical, and a hydroxyl anion. FeII is then reduced back to iron I by another round of reaction with hydrogen peroxide to a FeI peroxide radical, and a proton. The hydroxyl radicals oxidize DAB, producing a localized red-brown product. This technique has been used to study heme absorption in the rat (7) and dog (8) models. Rats, like many eukaryotes, synthesize heme in addition to taking it up from their diet. The purpose of this study was to follow the pathway of intestinal heme uptake in the rat duodenum. Electron micrographs showed that heme was released from hemoglobin in the intestinal lumen, then entered the mucosal cels through the apical pits. The heme was next observed in the apical tubules, and was finaly observed in the secondary lysosomes of the apical and supranuclear regions of the cel (7). However, the experimental conditions difered greatly from normal heme uptake by the rat. The heme was administered surgicaly to the animals in quantities sufficiently large enough se a reaction product generated by DAB (7). The canine model has been used to demonstrate that the sub-celular route of 28 hemoglobin?iron absorption through intestinal epithelial cels is diferent from the absorption of inorganic iron (8). Studies using this model concluded that endocytosis appeared to play a major role in heme absorption, offering insight into a possible route of hemoglobin-iron. In the current study, we have established a whole animal heme staining methodology using the natural heme auxotroph, C. elegans. By carefully controlling the heme levels in the growth medium of this nematode, we characterized heme localization in mutants which show aberant growth in response to heme and in wild type N2 worms. To our knowledge, this is the first time heme localization has been characterized in a whole animal. This established protocol wil enable us to further our current understanding of eukaryotic heme traficking by providing a conceptual framework for approaches such as Raman confocal microscopy. 29 Chapter 2: Materials and Methods Worm strains and growth medium The N2 wild-type strain of C. elegans was used for al experiments except where indicated. Mutant worm strains IQ911, IQ1068, IQ938, IQ731, IQ828 were obtained by Dr. Anita Rao. These mutants are capable of surviving in high levels of heme (> 800 ?M) that are otherwise toxic to wild-type worms. The worms were grown in axenic CeHR medium that was modified (mCeHR) to eliminate al sources of exogenously added heme under aerobic conditions at 20?C in tisue culture flasks. Synchronized L1 larvae were added to mCeHR axenic liquid medium at a population level of ~1500 worms/mL and grown in designated heme concentrations until they reached the mid-L4 larval stage, at which point they were fixed, permeabilized, and stained for heme. Worm permeabilization and fixation for light microscopy N2 worms were grown to the gravid stage in growth medium containing 20 ?M hemin chloride (Frontier Scientific, Logan, UT) and bleached [1.1 % Clorox bleach (sodium hypochlorite)/0.55 M NaOH] to harvest the eggs. Eggs were washed 2-3 times with M9 buffer (86 mM NaCl (Fisher, #S271-43), 42mM Na 2 HPO 4 (Fisher, #S774), 22mM KH 2 PO 4 (Acros, #42420-5000), 1mM MgSO 4 7H 2 O (Acros, #423905000)) and synchronized by hatching overnight in M9 bufer solution. When the worms reached the L4 larval stage, they were harvested for fixation and permeabilization. The L4 worms were washed thre times with M9 buffer 30 solution and then incubated in the M9 buffer solution for 1 to 3 h to empty gut contents. Afterwards, the animals were transfered to a 1.5 mL microfuge tube, alowed to setle for 10 min, washed with 1.0 mL distiled water and alowed to setle for an additional 10 min. The permeabilization method was modified from a previously published method (55). After the supernatant of the setled worms was aspirated, the microfuge tube was placed on ice. One ml of ice-cold paraformaldehyde solution [50 % Modified Ruvkin?s Witches Brew: (160 mM Potasium Chloride (Acros, #42409- 0010), 40 mM NaCl (Fisher Scientific, #S271-3), 20 mM EGTA-Sodium salt, 10 mM spermidine HCl (MP Biomedicals, #194542), 30 mM Pipes pH 7.4 (Acros, 17261- 1000), 50 % methanol (Fisher, #A412-4), 40 % distiled water, 2 % paraformaldehyde (Sigma, #41678-0010)] was added to the centrifuge tube containing the worms and the contents were mixed gently so to avoid damaging the worm tisues. The centrifuge tube was incubated at 4?C for 35 min with occasional mixing. Folowing this step, the worms were alowed to setle and washed twice with 1.0 mL Tris Triton Buffer [100 mM Tris HCl pH7.4 (Acros, #14050-0025), 1.0 % Triton X-100 (Acros, #422435-5000), and 1.0 mM EDTA (Acros, # 40997-5000)]. The animals were then incubated at room temperature in 1.0 mL of 1 % ?-mercaptoethanol (Acros, #12547- 1000) in TB for 15 min at eight RPM on a platform shaker (Platform shaker by Barnstead Thermolyne). The worms were washed once with 1X borate buffer [25mM H 3 BO 3 , 12.5mM NaOH)] and alowed to setle. One mL of 10 mM DL-1,4 Dithiothreitol 99 % (Acros, #16568-0050) in 0.9X borate buffer (22.5 mM H 3 BO 3, 11.25mM NaOH) was added to the tube of worms. The centrifuge tube was rotated 31 on the Labquake shaker for 15 min at room temperature. The worms were then rinsed five times with distiled water and stored overnight at 4?C in PBS buffer [13.7 mM NaCl (Fisher # 5271-3), 0.27 mM KCl (Acros, #42409-0010), 0.43 mM Na 2 HPO 4 (Acros, # 42437-5000), 0.14 mM KH 2 PO 4 (Acros, # 42420-5000) that had been deoxygenated with N 2 gas. These fixed and permeabilized worms were stained the following day. Pre-staining solution to eliminate oxygen radicals To eliminate oxygen radicals that can oxidize the staining chemical, 97 % 3- 3?diaminobenzidine tetrahydrochloride, 97 % (Acros, #112090050) solutions used in al experiments (unles otherwise noted) were deoxygenated by flushing with nitrogen gas for 15 - 20 min just prior to staining. Fixed and permeabilized worms were rinsed thre times with deoxygenated potasium phosphate buffer (pH 7.4) (4.05 mM K 2 HPO 4 and 0.95 mM KH 2 PO 4 ) incubated in the dark at 37?C for 2 hr in a solution containing 0.2 % bovine liver catalase (Sigma, # C-1345) (56), 0.04 % bovine liver superoxide dismutase from bovine liver (Sigma, #S-186) in potasium phosphate buffer, pH 7.4. The worms were then rinsed twice with deoxygenated 0.06 M Tris HCl bufer (pH 8.0-8.2). Heme staining for light microscopy A solution of 0.15 % DAB in 0.06 M Tris HCl (pH 8.0-8.2) was added to al samples. Then, H 2 O 2 (Sigma, #H1009) was added to each staining group for a final concentration of 0.06 %. Control samples were incubated with DAB in the absence 32 of H 2 O 2 . Al samples were incubated for 30 min under deoxygenated conditions and rinsed thre times with deoxygenated, distiled H 2 O. The worms were mounted on 0.1 %? 0.2 % agar (Fisher Bioreagents, # BP26411) beds, prepared on glas slides and photographed using a Leica DMIRE2 DIC microscope fited with a Retiga 1300 cooled 12-bit CD camera and imaged with SIMPLEPCI version 5.2 Software (Compix, Inc.). Heme staining analysis The control groups of worms were always evaluated before the experimental groups to ensure that staining in the controls was sufficiently minimal to permit analysis. Staining in the controls typicaly varied slightly in intensity, but the final reported percentage of staining in the controls included al variations of staining intensity. The majority of control groups in the following experiments had ~ 10 % staining. One or two slides containing worms from the control group was evaluated. Each slide contained ~300-500 worms. The entire slide was examined to determine the percent of worms stained out of the whole group. The experimental groups were evaluated in the same manner as the control group. While there was some variation in staining intensities in the experimental samples, final conclusions were based upon the degre to which > 70 % of the worms stained. Worms were examined and photographed using bright field light microscopy. Diferential interference contrast (DIC) microscopy was not used in the majority of the heme staining experiments due to dificulties capturing heme staining in photomicrographs. 33 DAPI staining An unsynchronized population of wild-type worms were grown at 20 ?M heme and fixed by the chemical fixation method described in this section. The worms were then stained with a 1 ?g/ml DAPI solution (Sigma, #32670) for 10 min at room temperature. Worm fixation and permeabilization using microwave fixation Due to poor preservation of tisue structure integrity, the fixation and permeabilization method used for light microscopy could not be used for electron microscopy. Instead, a C. elegans fixation and permeabilization protocol was modified from a previously described method (57) (communication, Dr. David Hal at the Albert Einstein College of Medicine). Wild?type N2 worms were grown in axenic mCeHR medium containing 80 ?M heme. When the worms reached mid-L4 stage, they were rinsed once with M9 buffer solution and incubated in the M9 buffer solution for 1 to 3 h to clear gut contents. The M9 buffer solution was changed once every hour. Worms were washed once with distiled H 2 O and placed on ice for transport to the University of Maryland Laboratory of Biological Ultrastructure (LBU) for microwave fixation (Pelco 3440 Laboratory Microwave oven and Pelco 3420 Microwave Load Cooler). At the LBU, the worms were rinsed thre times with fixative solution [1.0 % glutaraldehyde (Acros, # 233280250), 0.05 M sucrose (Acros, #220900010), and 0.1 M Hepes, pH 7.4. (Sigma, #H-3375)] and placed into one wel of a thre-wel glas chamber slide. The hotspots of the microwave were detected prior to each microwave fixation so that the worms were microwaved within 34 the hotspots. The hotspots were predetermined using a probe placed over the grid located on the floor of the microwave oven. The probe contained individual bulbs that aligned with each grid box. When the microwave was turned on, the bulbs that flashed red indicated the areas of the grid that contained the hotspots. These hotspots varied from use to use. The glas chamber slide was placed into an ice bath (a 150 m Petri dish full of crushed ice). The chamber wel containing the worms was placed directly over a previously determined hotspot. The worms were microwaved at 70?C at full power for 1.5 min ON and for 2 min OF for a total of five cycles. After each cycle, the glutaraldehyde fixative solution and the ice bath were changed. This proces was repeated five times. Once these five cycles were completed, the worms were microwaved at the same temperature and power setings for 1 min for thre cycles with a 4 min recovery period. The total time exposed to microwave energy was 10.5 min. The animals were then incubated at 37?C for 1 h in the 0.2 % catalase and 0.04 % superoxide dismutase solution in 5 mM phosphate buffer, pH 7.4., described earlier. The worms were rinsed twice with 0.06 M Tris HCl (pH 8.0- 8.2) and divided into control and staining groups. Heme staining for electron microscopy A solution of 0.15 % 3-3?diaminobenzidene, 0.06 M Tris HCl (pH 8.0-8.2), and 0.06 % H 2 O 2 was used to stain worms for 30 min at room temperature under deoxygenated conditions. The control group was incubated in the absence of hydrogen peroxide. After staining, the worms were rinsed thre times with deoxygenated, distiled water. The worms were then incubated in a counterstaining and fixative solution of 0.1 % 35 osmium tetroxide (EMS, 19140) for 33 min at room temperature. The worms were then washed five times with deoxygenated, distiled water and embedded in epoxy resin. Worm embeding Al materials used for embedding were purchased from Electron Microscopy Services (EMS, Fort Washington, PA) unles otherwise noted. Agarose beds were made by pouring 3.0 - 4.0 mL of 3.0 % Low Melt Preparative Grade Agarose (BioRad, 162- 0019) into 60 m petri dishes. Approximately four to twelve worms were aranged side by side on top of the first layer of the solidified agarose and then covered with another 3.0 to 4.0 mL of agarose, so the worms were sandwiched betwen the two beds. The Petri dishes were stored overnight at 4.0?C with a few drops of water on top of the agar beds, to prevent the agar from drying out. The next morning, each group of worms were cut out of the bed with a razor blade to a size that would fit into the tips of the embedding molds (Pelco International, #105). Each piece of agarose contained four to twelve worms that could be placed coronaly or sagitaly into the tips of the mold. The pieces of the agarose beds could not be bigger than the mold nor smaler than the tip of a Pasteur pipete. The worm samples in agarose were dehydrated by placing them in 25 mL glas bottles and incubating for 5 min in each of the following ethanol grades: 30 %, 50 %, 75 %, and 95 %. For the final ethanol dehydration step, the worms were incubated in 100 % ethanol thre times, each time for 10 min. Further dehydration steps were necesary, and they were caried out in propylene oxide (PO) (1,2- 36 Epoxypropane, Methyloxirane, (#20412). Al PO steps were conducted using only glas or polypropylene equipment, as PO is highly reactive. The agarose pieces were incubated in PO thre times, each time for 10 min. Worms were incubated in a 2:1 ratio of PO:resin mixture for 2 to 3 h on a vertical rotator (Barnstead Thermolyne) at room temperature. The resin mixture contained 45.0 % Embed 812 (#14900), 30.0 % Dodecenyl Succinic Anhydride (DSA) (#13710), 25.0 % Nadic Methyl Anhydride (Methyl-5-Norbornene-2,3-Dicarboxylic Anhydride) (#19000), 1.5 % DMP-30 (2,4,6-Tri (dimethylaminomethyl) phenol, (#13600). Resin was made fresh for each experiment and stored under vacuum in a room temperature desicator. The worm sections were then incubated in a 1:2 PO:resin solution overnight at room temperature on the vertical rotator. Approximately 15 h later, the agarose pieces were removed from the PO: resin mixture and placed into resin for 3 h at room temperature. During this 3 h incubation period, the resin was changed twice. Following the second change of resin, the agarose pieces were positioned at the tip of the embedding mold with the worms either coronaly or sagitaly placed. Smal pieces of paper with the asigned sample code, which diferentiated the samples from one another, were placed into the molds. Resin was poured into the embedding molds, making sure the worms remained correctly aligned at the tip. The molds were cured for 3 days at 60?C, removed from the oven, and alowed to cool for at least half a day. Thin sectioning At the end of the embedding proces the worms were thin-sectioned for electron microscopy (EM) on an American Optical Ultracut microtome with a glas knife 37 (Ultramicrotomy grade glas (# 71012) generated by a glas knife cutter (LKB 7800B Knifemaker). Water boats to catch the thin sections, were made with plastic water boats (#71007) and sealed to the glas knife with clear nail polish (#72180). A chloroform solution was used to flaten the sections and a Zerostat Ion Gun (153) was used to minimize static electricity. The sections were approximately 60 nm ? 90 nm thick, and were placed on coated 375?mesh copper grids. The samples on the grids were counterstained with lead citrate (#17800) and uranyl acetate (#22400). The uranyl acetate solution was made by disolving 0.2 g dry uranyl acetate in 10 mL of distiled H 2 O. This solution was stired overnight at room temperature. The lead citrate solution was made by disolving 0.03 g in 10 mL H 2 O and adding 0.1 mL 10 M NaOH, shaking vigorously by hand for 5 min and storing overnight at 4?C. Prior to staining, these solutions were centrifuged for 10 min at room temperature to remove any particulate mater. Electron micrographs were taken under the direction of Mr. Tim Maugel, with a Zeis EM10 Transmision Electron Microscope at the LBU. 38 Chapter 3 Characterization of heme localization in wild- type worms using light microscopy Sumary Heme is a tetrapyrrole esential to life for many organisms. In most of these organisms, heme is both synthesized in a wel-conserved eight-step pathway, as wel as absorbed from the diet. Heme uptake has been studied morphologicaly since the 1960?s, but most of the animals used for those studies were heme synthesis prototrophs. We use the nematode Caenorhabditis elegans as the model organism of choice to study heme transport because we recently discovered that it does not synthesize heme but utilizes heme from diet (13). Thus, the in situ study of heme uptake in this animal wil provide a beter understanding of how heme transport is mediated in eukaryotes without interference from endogenous heme sources. Here, we show that, by using a DAB staining technique and light microscopy, C. elegans stains robustly for heme, revealing a direct corelation betwen histochemical heme staining and the level of nutritious heme administered exogenously in the growth media. Wild-type worms that were grown at 1.5 ?M, 4.0 ?M, 20 ?M, 40 ?M, and 100 ?M heme showed a proportional increase in the intensity of heme staining. Our systematic standardization of the methodology of heme staining builds upon earlier work, and establishes an in vivo approach to directly visualize heme in an intact whole animal. This new method is a useful starting point for the study of in situ heme traficking in C. elegans. 39 Rationale This research project was designed to cytologicaly localize and characterize heme staining in C. elegans using microscopy. The overal research goal of our research group is to discover the molecules controlling heme transport pathways in C. elegans. C. elegans is a fre-living nematode that does not synthesize heme, but requires exogenous heme for incorporation into hemoproteins (13). We hypothesize that specific intracelular pathways must exist for heme transport because heme is an esential molecule that cannot simply difuse into, within and betwen cels due to its hydrophobicity and cytotoxicity. C. elegans is an ideal organism to addres questions pertaining to heme transport for a number of reasons. First, C. elegans does not contain any of the wel-conserved genes required for heme biosynthesis, and thus we have complete control over heme transport in the animal. Second, C elegans is translucent, proving to be an ideal system to localize heme within cels and organs in the intact animal. Third, C. elegans is a wel-established genetic model with invariant somatic cel number. It has been used for genetic and cel biological studies for more than thirty-five years (58). Complete control of heme transport is dificult to atain in eukaryotic organisms that synthesize heme (13). Establishing a protocol for in situ heme labeling using the chromogenic substrate 3-3? diaminobenzidine (DAB) wil alow us to gain a beter understanding of where hemes are localized and transported in vivo in C. elegans, a natural heme auxotroph. Histochemical staining with DAB exploits the peroxidase activity of heme, and has been used extensively for the last forty years to 40 study heme uptake and traficking in various model organisms. When hydrogen peroxide is added as a second substrate, the iron in the center of the heme ring reacts with the hydrogen peroxide, releasing oxygen radicals. These oxygen radicals oxidize DAB and produce a localized brown/red reaction product (17). Previous heme labeling microscopy studies in other organisms used artificialy elevated heme levels in organ slices. The studies described herein, however, localize histochemical heme in situ in intact C. elegans under normal and varying levels of physiological heme concentrations. Results Our goal for this project was to cytologicaly localize heme in whole animals using C. elegans. This manner of cytologicaly localizing heme in worms has not been reported in the literature. As a starting point, we conducted our first staining experiment by atempting to use a procedure established by Graham and Karnovsky (17). This study described a new histochemical staining technique using DAB, to study horseradish peroxidase absorption in the mamalian kidney, using the rat model. This method involved incubating the worms for 10 min in a DAB-saturated staining solution in 0.05M Tris HCl buffer, pH 7.6. Two groups of worms were tested with this staining procedure. The first group of worms had been fixed and permeabilized two days earlier as described in the Materials and Methods section (59). The second group of worms was fixed and permeabilized by the same method, on the same day. Within 10 min of the staining reaction, the worms in the control samples (minus H 2 O 2 ) , permeabilized two days earlier, showed non-specific staining. 41 However, no staining appeared in either the control group (minus H 2 O 2 ) or the experimental group (plus H 2 O 2 ) of the worms that had been fixed on the same day. After these results, we decided to try to modify the protocol as described by Simionescu et al., that was used for one of the first EM studies for heme uptake in mamalian intestine (60). This modification caled for a staining incubation period of 60 min and for the DAB Tris-HCl buffer solution to have a pH of 8.0. This experiment produced significant levels of non-specific staining. We then decided to try to establish our own heme staining methodology. Adition of catalase to pre-incubation solution Our first heme staining experiment, based on a previously published method (17) produced inconclusive results. There was no diference in staining betwen the worms in the control samples and worms in the experimental samples. Our first thought was that endogenous hydrogen peroxide could be oxidizing DAB, producing non-specific staining in the control samples. To eliminate this endogenous hydrogen peroxide, we incubated the worms, post-fixation, in a 0.2 % catalase (CAT) solution in 5 mM phosphate buffer for 30 min at 37?C, as described in the Materials and Methods section. CAT is the enzymatic catalyst in the chemical reaction below: CAT 2 H 2 O 2 -->2 H 2 O + O 2 While there was stil significant non-specific staining, we noticed for the first time that gravid worms contained the majority of the non-specific staining. Staining was reduced in the samples by ~ 40 % (Figure 3). 42 Figure 3: Adition of CAT to eradicate peroxide radicals. A mixed population of N2 worms grown at 20 ?M heme in mCeHR medium was permeabilized as described in Materials and Methods and stored overnight in PBS at 4?C. After 15 h, the worms were incubated in 5 mM potasium phosphate buffer, pH 7.4 containing 0.2 % CAT for 30 min at 37?C, in the dark. The staining reaction was caried out for 30 min at room temperature. Staining in the control samples incubated with CAT by ~ 40 %. This experiment was performed one time, and each micrograph is a representative image. Scale bar = 100 ?m. -CAT +CAT -H 2 O 2 +H 2 O 2 43 44 While we achieved a significant improvement in the non-specific staining of the controls by adding the CAT step, we felt it was necesary to further standardize the staining procedure. We tested thre additional concentrations of CAT for a longer period of time to se if there would be any diferences in the staining of our control samples. In addition to the 0.2 % used originaly, we tried 0.02 %, 0.1 %, 0.2 % and 1.0 % of CAT in 5mM phosphate buffer, pH 7.4 for 2 h at 37?C. No diferences were detected in staining of the controls betwen these four concentrations. The experimental samples of gravid worms in each of the diferent concentrations of CAT stained only 30 - 40 % greater than the gravid worms in the control samples. We again noted that the gravid worms in the control samples had non-specific staining while the non-gravid adults and larval worms in the control samples had very litle, if any, non-specific staining. Experiments conducted later confirmed that larval worms consistently had les non-specific staining than gravid worms. Adition of SOD to pre-incubation solution. Although we significantly reduced the non-specific staining in our control samples by incubating the worms with CAT, we added another modification. Since endogenous H 2 O 2 is generated by the reaction catalyzed by the enzyme superoxide dismutase (SOD), we rationalized that pre-incubation with SOD would help reduce the non- specific staining observed in the control samples. SOD catalyzes the following reaction in conjunction with CAT. ! O 2 " +O 2 " +2H + # SOD O 2 +H 2 O 2 ! 2H 2 O 2 " CAT 2H 2 O+ 2 45 Surprisingly, > 90 % of the worms in the experimental group stained for heme, while < 5 % of the control group revealed non-specific staining. The staining observed in the experimental group was throughout the entire worm and the staining intensity was uniform, as shown in Figure 4. No other distinct staining paterns were observed. Heme staining unsynchronized worms grown at 20 ?M and 100 ?M heme The addition of SOD and CAT to our protocol showed promising results by reducing the non-specific staining. We then decided to try and compare staining levels betwen two mixed, unsynchronized populations of worms grown at 100 ?M and 20 ?M heme. The worms were fixed, permeabilized, and 15 h later, pre-incubated with 0.2 % SOD and 0.2 % CAT in 5 mM phosphate bufer, pH 7.4. The worms were then stained for heme with 0.15 % DAB in 0.06 M Tris HCl, pH 8.0. The results from this experiment are shown in Table 1. During this experiment there were no diferences in staining betwen the gravid worms, and their embryos, in the control and experimental groups. DAPI staining of the nucleus in these worms, described in Materials and Methods, revealed that >75 % of the eggs and 90 % of worms were permeabilized. Control samples of worms grown at 20 ?M and 100 ?M heme did not difer in their staining intensity. Importantly, worms grown at 100 ?M heme from the experimental group had significantly darker and more intense heme staining than experimental worms grown at 20 ?M heme. This experiment indicated that worms grown in diferent concentrations of heme display diferences in the intensity of heme staining. These results also indicated that gravid worms have significantly higher levels of non-specific staining, independent of the addition of H 2 O 2 . 46 Figure 4: Adition of SOD to eradicate oxygen radicals. A mixed population of wild-type worms was grown in 20 ?M hemin and their F 1 progeny were synchronized as described in Materials and Methods. The L1 larvae were inoculated into 20 ?M heme in mCeHR medium. At the L4 larval stage the worms were harvested, fixed, permeabilized, and stored overnight at room temperature. The following morning the worms were incubated in a pre-incubation solution of 0.2 % CAT and 0.2 % SOD in 5 mM phosphate buffer for 1 h. The worms were subsequently stained for heme. Les than 5.0 % of the control samples stained while > 90 % of the experimental samples stained. This experiment was performed one time and each micrograph is a representative image. Scale bar = 100 ?m. -H 2 O 2 +H 2 O 2 47 48 Table 1: Larval worms have significantly les non-specific staining than gravid worms. The majority of worms in the control group that stained were gravid. Approximately 5 % of the larvae in the control samples had non-specific staining. Greater than 85 % of the stained worms in the experimental group, however, stained specificaly for heme. [Heme] Control Experimental Gravids Larvae Gravids Larvae 20?M ~40 % 5 % > 90 % > 85 % 100? ~40 % 5 % > 90 % > 90 % 49 To eliminate the problem of non-specific staining in gravid worms, we conducted our next experiment using only larvae. This experiment was also designed to further test the possibility of diferential heme staining of worms grown at diferent heme concentrations. Heme staining in synchronized L4 larvae at 20 ?M and 100 ?M heme To test whether worms that were grown in higher heme concentrations stained more intensely than worms grown at lower heme levels, synchronized L1 larvae were grown in mCeHR medium in 20 ?M and 100 ?M heme. The results are shown in Table 2 and Figure 5. Worms grown in 100 ?? heme stained approximately 2- 3 fold more intensely than worms grown in 20 ?? heme as shown in Figure 5. Synchronized L4 larvae grown at 1.5 ?M, 4 ?M, 20 ?M, and 40 ?M heme were stained for heme using deoxygenated conditions. While we significantly improved the specific staining with the addition of SOD and CAT, and using only larval worms, we observed that worms in the control group would begin to stain during the rinsing proces, post staining. Because staining results from the oxidation of DAB, it was possible that oxygenation of the rinsing solution was causing this experimental artifact. We decided to go one step further and stain the worms, now under deoxygenated conditions, using the apparatus shown in Figure 6. The results are shown in Table 3 and Figure 7. As might have been predicted, worms grown in 40 ?M heme stained the most intensely. The worms grown in 20 ?M heme concentrations stained slightly more intensely than the L4 larvae grown at 4 ?M heme. 50 Table 2: Synchronized L4 larvae exhibit specific heme staining. Approximately 5 % of the 100 ?M larvae revealed staining throughout the body in the control group. Worms grown at 100 ?? heme showed > 95 % heme staining in the experimental group and > 80 % of those worms showed uniform staining throughout the body. Worms grown at 20 ?? showed staining in > 80 % of the experimental group and this staining was throughout their body. Les than 5 % of the 20 ?M worms stained in the control group. [Heme] Control Experimental 20 ?M < 5 % > 95 % 100 ?M ~ 5 % > 80 % 51 Figure 5: Synchronized L4 larvae grown at 20 ?M and 100 ?M heme exhibit diferences in heme staining intensity. Synchronized L1 larvae obtained from N2 worms grown in 20 ?M heme were inoculated into mCeHR medium containing either 20 ?M or 100 ?M heme. When these larvae reached the L4 stage, the worms were fixed, permeabilized, and stored overnight as described in Materials and Methods. Approximately 15 h later, the worms were stained for heme. The experimental samples from the worms grown at 100 ?M heme stained 2 -3 fold more intensely than the worms grown at 20 ?M heme. This experiment was performed one time and each micrograph is a representative image. Scale bar = 100 ?m. -H 2 O 2 +H 2 O 2 20 ?M 10 ?M 52 53 Figure 6: Experimental apparatus for deoxygenation of worm samples. Each of the worm samples were contained in a plastic test tube fited with a rubber septum. The needle of the apparatus was the source of N2 gas for the test tubes containing the samples. A separate needle was inserted through the rubber septum of the test tubes, alowing the oxygen and nitrogen to escape. Tubing connected to nitrogen gas tank Needle attachment for sample tubes fitted with a rubber septum (see image below) 54 Needle allows oxygen to escape Test tube containing sample Needle attachment (through which the N 2 gas flows) 55 Table 3: Syncronized L4 larvae exhibit significant levels of specific heme staining when stained under deoxygenated conditions. Synchronized L1 larvae from wild-type worms grown in mCeHR medium and 20 ?M heme were inoculated in mCeHR medium, supplemented with 1.5 ?M, 4 ?M, 20 ?M, and 40 ?M heme until they reached the L4 stage. The synchronized L4s were fixed, permeabilized, and stored overnight as described in Materials and Methods. Approximately 15 h later, the worms were stained for heme under deoxygenated conditions described in Materials and Methods. Al experimental samples had > 80 % specific heme staining. [heme] Percent of worms that stained within each group Control Experimental 1.5 ?M 0 % > 80 % 4.0 ? 0 % > 80 % 20 ?M ~ 5 % > 80 % 40 ? 15 % ~ 90 % 56 Figure 7: Synchronized L4 larvae grown at 1.5 ?M, 4 ?M, 20 ?M, and 40 ?M heme and stained for heme under deoxygenated conditions. Wild-type worms were grown in 20 ?M heme and mCeHR medium. Their eggs were harvested by bleaching and hatched overnight in M9 buffer as described in Materials and Methods. The synchronized L1 larvae were then inoculated in mCeHR medium supplemented with 1.5 ?M, 4 ?M, 20 ?M, and 40 ?M heme and alowed to develop to the L4 stage. The synchronized L4 worms were fixed, permeabilized, and stored overnight as described in Materials and Methods. The larvae grown at 40 ?M heme had the most intense staining of al the experimental samples, while the diference in heme staining intensity betwen 1.5 ?M, 4 ?M, and 20 ?M was not significant. This experiment was performed one time and each micrograph is a representative image. Scale bar = 100 ?m. -H 2 O 2 +H 2 O 2 1.5 ?M 4 ?M 20 ?M 40 ?M 57 58 The experimental samples from the worms grown at 1.5 ?? heme exhibited staining that was not significantly diferent from staining of experimental worm samples from 4 ?M and 20 ?M heme. The diference in staining intensity betwen 4 ?M, and 20 ?M larvae was not as great as the diference betwen either of those groups and 40 ?M (Figure 7). Mixed population of worms grown at 1.5 ?M, 4 ?M, 20 ?M, and 40 ?M heme We were able to drasticaly reduce the non-specific staining in our control samples by incubating the worms in a CAT/SOD solution and staining the worms under deoxygenated conditions. We then decided to stain mixed populations of worms to determine if deoxygenated conditions would reduce the non-specific staining in gravids in the control groups. Mixed populations of worms were grown in 1.5 ?M, 4 ?M, 20 ?M, or 40 ?M heme, then fixed, permeabilized, and stored overnight as described in Materials and Methods. The worms were then stained for heme as described in Materials and Methods. As shown in Table 4, > 80.0 % of gravid worms in the control group stained, and the larvae in this group showed minimal levels of non-specific staining. Based on these results and results from previous experiments, we decided to use synchronized non-gravid worms for future staining. To determine the optimal time for analysis, the stained worms were analyzed either imediately or the following day. Optimal results were obtained when samples were analyzed the same day the staining procedure was performed. 59 Table 4: Diferences in specific heme staining betwen larval and gravid worms at 1.5 ?M, 4 ?M, 20 ?M, and 40 ?M heme. Mixed populations of worms were grown in mCeHR medium and 1.5 ?M, 4 ?M, 20 ?M, or 40 ?M heme. They were permeabilized, fixed, and stained for heme as described in Materials and Methods. For gravid worms, there was no diference in staining betwen experimental and control groups. Larval worms, however, exhibited significant levels of specific heme staining. Percentage of worms that stained within each sample [heme] Control Experimental Gravids Larvae Gravids Larvae 1.5 ?M > 80 % < 1 % 100 % ~ 80 % 4 ?M 100 % ~ 5 % 100 % ~ 70 % 20 ?M > 90 % < 1 % 100 % ~ 60 % 40 ?M > 90 % < 10 % 100 % ~ 80 % 60 Standardization of necesary SOD concentration In order to determine the optimal staining conditions, we used varying levels of SOD (0.02 %, 0.06 %, 0.1 %, 0.15 % and 0.2 %). While we were trying to establish the minimum concentration of SOD required to minimize staining in our control samples, there were additional results from this experiment. We were able to determine that a concentration of 0.02 % SOD sufficiently minimized non-specific staining; les than 10 % of the worms in the control samples stained. We also observed, for the first time, distinct, punctate intestinal heme staining in the experimental groups. Two punctate staining paterns were observed. Punctate staining sen outside the intestine (Figure 8A and 8B) or in the intestine (Figure 8C and 8D). One other observation we noted throughout al of our heme staining experiments was that the majority of heme staining was concentrated in the anterior half (first 5 pairs) of intestinal cels (Figure 9). 61 Figure 8: Intestinal punctate staining. Wild-type worms were grown in 20 ?M heme in axenic mCeHR medium. Their progeny was synchronized and grown in mCeHR medium containing 4 ?M heme. The worms were harvested at the mid-L4 stage, fixed, permeabilized, and stored overnight as described in Materials and Methods. Approximately 15 -18 h later, the worms were stained for heme as described in Materials and Methods. Two punctate heme staining paterns were observed. One patern showed heme staining outside the intestine (A and B), while the other patern exhibited heme staining within the intestinal cels (C and D). The arows within each of these figures indicate the punctate staining. This experiment was performed one time and each micrograph is a representative image. A, C, E scale bar = 100 ?m. B and D scale bar = 25?m. -H 2 O 2 A B C D 62 A B + H 2 O 2 + H 2 O 2 E 63 Figure 9: Heme absorption in intestinal cels. Larval worms were grown in 1.5 ?M heme, harvested, fixed, and permeabilized as described in Materials and Methods. Approximately 15 h later, these worms were stained for heme as described in Materials and Methods. The heme staining signal was most intense in the anterior intestinal cels and decreases in intensity until there is almost no detectable signal in the posterior intestinal cels. Scale bar = 100 ?m. -H 2 O 2 +H 2 O 2 head tail 64 head 65 Chapter 4: Aplication of new heme staining technique Sumary We have developed and optimized a methodology to histochemicaly detect in situ heme under normal physiological conditions in C. elegans. We applied this heme staining protocol in thre capacities: to evaluate qualitative diferences in heme levels and localization betwen the mutant and wild-type worms, to explore the possibility that the maternal heme efect in C. elegans could be visualized by staining, and to continue heme staining studies of wild-type worms at the electron microscopy (EM) level. Our research group identified 13 strains of C. elegans mutant in heme homeostasis and 5 of these mutant strains were selected for this heme staining study. We show that al but one of the five mutants selected exhibit les intense heme staining than wild-type. We then tested possible heme staining diferences caused by the maternal heme efect. A maternal efect occurs when the offspring phenotype is due solely to substances directly pased to the offspring from the mother, whether the substances are mRNA, proteins, or other molecules, such as heme. Results from this experiment show that C. elegans progeny of wild-type worms grown in diferent heme concentrations stain diferently for heme. Finaly, to more precisely locate heme within individual cels, we applied our heme staining technique to EM experiments in order to characterize heme staining at the ultrastructure level in C. elegans. In each of these experiments, we experienced isues with non-specific staining in our control samples, preventing us from drawing any insightful 66 conclusions about heme localization in C. elegans at the EM level. However, these results also indicated that heme staining C. elegans at the level of the light microscope provides a starting point from which to examine heme uptake and localization at the ultrastructure level. Rationale We hypothesize that heme homeostasis is mediated by specific molecules. To identify the genetic determinants of heme, a forward genetic scren conducted in our research group obtained thirten strains of worms mutated in heme homeostasis. Phenotypicaly, these worms are capable of surviving in 800 ?M or 1 mM concentrations of heme, levels that are toxic to wild-type worms. In principle, these mutants could difer from wild-type worms in that the heme transport pathway(s) either transport les heme into intestinal cels, sequester heme more eficiently, or more rapidly detoxify heme. In order to further understand heme transport in C. elegans, we utilized our heme staining technique and compared diferences in heme levels betwen the mutant and wild-type worms grown at low concentrations of heme. A maternal efect occurs when the source of an offspring phenotype is due solely to substances including mRNA, androgens or other compounds that are pased on to the offspring by the mother. This phenomenon has been characterized in wasps, birds, and humans (61-64) and most importantly, in C. elegans, involving proteins such as phosphatase, isomerase, and CDK (65,66). C. elegans requires heme for growth and development but is unable to synthesize it. The fact that transfering 67 maternal substances to the progeny is vital to their embryonic growth and development led us to postulate that maternal heme might also be pased on to progeny. Furthermore, while C. elegans is a non-parasitic nematode, there are several known parasitic nematodes that are also heme auxotrophs. It is posible that the maternal proteins responsible for transfering heme to the embryos could potentialy be drug targets to limit these parasitic infections (13). Unpublished data have shown that when F 1 larvae obtained from P 0 worms that were grown in low (4 ?M), optimal (20 ?M), and high (500 ?M) heme were inoculated into medium containing no added heme, F 1 progeny of the worms grown at 4 ?M heme growth arested at the L3 larval stage, while the F 1 progeny of the worms grown at 500 ?M growth arested as early adults. These results suggested that heme levels available to the mother afected the subsequent growth and development of her ofspring. We therefore examined diferences in heme staining intensity and localization betwen the progeny of worms grown at 4 ?M, 20 ?M, and 500 ?M heme. Previous work conducted to study eukaryotic heme uptake and localization at the ultrastructure level used abnormal, non-physiological conditions of heme-loading in organisms that synthesize heme. One of these studies designed to characterize heme uptake in the rat duodenum at the EM level used heme administered to rats as hemoglobin or hemin chloride in highly concentrated quantities through an intragastric tube (7). Administration of high quantities of hemoglobin/hemin chloride is abnormal as wel as non-physiological. We have standardized a technique to stain heme in whole animals, under normal growth conditions of heme, using the heme 68 auxotroph C. elegans. It is our goal to apply this heme staining technique in EM experiments in order to characterize heme localization in wild-type worms. Using C. elegans alows us to completely control the concentration of heme available to the animal. Additionaly, the location of every one of its 959 somatic cels is the same in every animal, alowing us to precisely identify heme localization in each cel. Furthermore, the entire worm has been sectioned and the images are available online at ww.ormiage.org to use as a reference for characterizing where celular heme is located. Results Worms mutated in heme homeostasis display diferential heme staining as compared to wild-type worms. An unpublished study was conducted using random EMS-based mutagenesis in C. elegans to identify mutants resistant to toxic heme levels (800 ?M). Dr. Anita Rao isolated 13 mutants which belong to 5 complementation groups (Table 5). The rationale for characterizing these mutants for heme staining is that disruption in heme homeostasis by either heme transport or heme sequestration would lead to survival under toxic heme levels. Thus, the logical extension for our studies would be to examine how much and where heme localized in these mutants at the level of the light microscope. For this experiment, we analyzed one mutant from each complementation group, as indicated in Table 5. The experiment was conducted twice and the results are summarized in Tables 6 and 7. There was no staining observed in worms from the control groups (Table 6). Only 50 % of the worms in any of the experimental groups revealed heme staining. Al of the heme staining was 69 punctate, and restricted to the first 5 pairs of intestinal cels. Mutant worms IQ731, IQ938, IQ828, IQ911 had les intense heme staining than wild-type. However, IQ911 revealed the lowest intensity in the punctate staining than any of the other mutants (Figure 10). In contrast to the other four mutants, IQ1068 revealed an increase in staining intensity compared to wild-type worms. Progeny obtained from worms grown at diferent heme levels stain diferentialy for heme. Wild-type worms were grown for at least two generations in 4 ??, 20 ??, and? 500 ?? heme. Their eggs were harvested, synchronized overnight in M9 buffer. The following morning, the synchronized larvae were inoculated into mCeHR medium with no added heme, and alowed to grow for six days. On day 6, there were no visible gravid worms by microscopic examination in each of the thre groups. However, there were notable diferences in their developmental stage. F 1 progeny from 4 ?M heme were growth arested at the L3 stage while F 1 progeny from the 500 ?M heme were arested at the early adult stage. F 1 progeny from 20 ?M growth arested at the early L4 stage. The diference in growth rates of these worms is summarized in Table 8. The progeny were then histochemicaly stained for heme as described in the Materials and Methods section. No staining was observed in any of the samples in the control groups. The larvae that stained the most intensely for heme were from the progeny of worms grown at 500 ?M heme (Figure 11). The larvae from worms grown at 20 ?? heme stained les than half as intensely as the progeny of the worms grown at 500 ?M heme. Heme staining in larvae obtained from worms grown at 4 ?? heme was barely detectable, as shown in Figure 11. 70 Table 5: Complementation groups of heme-resistant mutant worms. The thirten mutants resistant to high heme are grouped in 5 complementation groups [courtesy of Dr. A. Rao and B. Le]. Mutants selected to analyze by heme staining are highlighted with a gray box. Complementation groups Mutants in each complementation group I IQ911 I IQ828 II IQ728 IQ938 IQ968 IV IQ921 IQ718 IQ731 V IQ1058 IQ1048 IQ1068d IQ1031 IQ1068 71 Table 6: Percent of specific heme staining in wild-type worms and worms mutated in heme homeostasis. N2, IQ938, IQ731, IQ911, IQ828 and IQ1068 worms were grown in mCeHR medium with 4 ?M heme. F 1 embryos were harvested, synchronized, and inoculated at 4 ?M heme. When they reached the mid-L4 stage, they were stained for heme as described in Materials and Methods. There was no staining in worms from the control groups. In worms from the experimental groups, only 50 % of the worms showed heme staining, which was punctate. Worm strain Control Experimental N2 0 % 50 % IQ938 0 50 IQ731 0 % 50 % IQ911 0 50 IQ828 0 % 50 % IQ1068 0 50 72 Table 7: Heme resistant mutants reveal diferential heme staining. The mutant worm strains IQ828, IQ938, and IQ938 al exhibited staining that was slightly les intense than wild-type, while IQ911 heme staining was significantly les intense than wild-type. Only one strain, IQ1068, exhibited heme staining that was more intense than wild-type. Mutant complementation group Relative intensity compared betwen wild-type and mutant I N2 > IQ911 I N2 ? IQ828 II N2 ? IQ938 IV N2 ? IQ731 V N2 ? IQ1068 73 Figure 10: Heme staining wild-type, IQ731, IQ911, IQ828, IQ1068, and IQ938 strains grown at 4 ?M heme. Wild-type, and mutant strains IQ731, IQ911, IQ828, IQ1068, and IQ938 were grown in 4 ?? heme in mCeHR medium for two generations. Their eggs were harvested and hatched overnight in M9 buffer. The synchronized L1 larvae were inoculated into 4 ?M ?heme in mCeHR medium. When the worms reached the mid-L4 stage, they were fixed, permeabilized, and stored overnight as described in the Materials and Methods section. Approximately 15 ? 18 h later, the worms were stained for heme using the protocol described in the Materials and Methods section. This experiment was performed two times and each micrograph is a representative image. Scale bar = 100?m . 74 -H 2 O 2 IQ828 IQ91 IQ731 +H 2 O 2 N2 75 +H 2 O 2 IQ1068 IQ938 -H 2 O 2 76 Table 8: Maternal heme efect on growth rates of progeny obtained from parental worms grown at diferent heme concentrations. F 1 progeny of worms grown at 4 ?M, 20?M, and 500?M heme were grown in mCeHR medium with no added heme for six days. After 24 h, the F 1 progeny of worms grown at 4 ?M heme had reached the L3 larval stage, the stage at which their growth was arested. F 1 progeny of worms grown at 20 ?M heme reached the mid-L4 stage after 49 h, after which they developed no further. The F 1 progeny of worms grown at 500 ?M heme reached the early adult stage after 94 h and did not ever become gravid. ND = not detected. [Heme] available to P 0 worms Time required for F 1 progeny to reach growth stages L3 Early L4 Mid-L4 Early adult 4?? 24 h ND ND ND 20?? 24 h 40 h 49 h ND 500? 24 h 40 h 49 h 94 h 77 Figure 11: Progeny obtained from worms grown at diferent heme levels reveal a maternal heme efect. Wild-type worms were grown for at least two generations in 4 ?M, 20 ?M, or 500 ?M heme. Gravid worms were bleached, and their eggs were harvested and synchronized overnight in M9 bufer as described in Materials and Methods. Synchronized L1 larvae were inoculated into separate flasks of mCeHR medium with no added heme. After 6 days, the worms were harvested, fixed permeabilized, and stained as described in Materials and Methods. Heme staining in larvae obtained from worms grown at 4 ?? heme was barely detectable. The larvae from worms grown at 20 ?? heme stained les than half as intensely as the progeny of the worms grown at 500 ?M heme. The larvae that stained the most intensely for heme were from the progeny of worms grown at 500 ?M heme. The arow bars indicate heme staining. This experiment was performed five times and each micrograph is a representative image. Scale bar = 100 ?m. head head head +H 2 O 2 -H 2 O 2 4 ?M 20 ?M 50 ?M 78 79 Preliminary results of heme staining C. elegans at the ultrastructure level For the electron microscopy studies we needed to modify the permeabilization and fixation protocol. C. elegans has a cuticle that is tough to penetrate by standard fixation and permeabilization methods. One permeabilization and fixation method available for C. elegans which alows for preservation of tisue integrity at the ultrastructure level is microwave fixation (57) Personal instructions were provided by Dr. David Hal, the Director of the Center for C. elegans Anatomy at the Albert Einstein College of Medicine. Microwave energy increases the rate at which fixative solutions penetrate the cuticle and subsequent underlying tisues. Additionaly, this fixation method improves the penetration of resin, the embedding material, without any further exposure of the sample to the microwave energy (57). The previous chemical method provided sufficient permeabilization of the cuticle for heme staining analysis at the light microscopy level. However, preservation of the tisue structure was insufficient for analysis at the ultrastructural level (Figure 12). We standardized the microwave fixation protocol in order to sufficiently preserve subcelular structures while simultaneously ensuring that heme staining can be observed at the ultrastructure level in C. elegans. Modifications were made to a previously published protocol (57) with personal communication from Dr. Hal. DAB is the chromogenic substrate of choice because it has a distinct, localized electron dense reaction product. Furthermore, it has been shown that the reaction product of the compound may not be afected by the embedding proces (17). It is important to note that we needed to re-establish the appropriate 80 concentration of DAB for microwave fixation because microwave fixation is not as penetrative as the chemical fixation that was used previously. Our first experiment tested thre concentrations of DAB: 0.15 %, 0. 30 %, and 0.45 %. We stained synchronized L4 larvae that had been grown at 20 ?M heme in mCeHR medium. The best results were obtained with worms with 0.15 % DAB, although there was very light staining in ~ 30.0 % of the worms from the control group. The two higher concentrations of DAB produced high staining (>50.0 %) in the control group. The tisue structure preservation, however, was superior to the previously used chemical fixation and permeabilization method as shown in Figure 12. We were satisfied with the preservation of tisue structure when observed under a light microscope, and proceded with our EM studies. Once we established the microwave fixation and heme staining protocols, we conducted a few electron microscopy studies to standardize the dehydration and embedding method given to us by Dr. Hal. As a first step, we decided to do the EM experiments with wild-type worms grown to mid-L4 stage in 80 ?M heme. We chose to stain worms grown at 80 ?M heme because we reasoned we would be able to se a greater heme staining signal without subjecting the worms to physiological toxic levels of heme. Mid-L4 larvae were harvested, fixed, and permeabilized by microwave fixation as described in the Materials and Methods section, and stained for heme using the DAB staining protocol previously standardized by us for light microscopy. This was followed by further fixation and counterstaining by osmium tetroxide, also described in Materials and Methods. The worms were embedded in 81 Figure 12: Comparison of microwave fixation to chemical fixation. Synchronized wild-type L1 larvae were grown in 20 ?M heme until they reached the mid-L4 stage. The L4 larvae were harvested, fixed, and permeabilized by microwave fixation as described in the Materials and Methods section. The same day, these worms were stained for heme. Permeabilization and fixation by microwave fixation showed substantialy improved celular structure preservation as compared to earlier studies using chemical fixation. This experiment was performed one time and each micrograph is a representative image. Scale bar = 100 ?m -H 2 O 2 +H 2 O 2 Chemical fixation Microwave fixation 82 83 epoxy resin, thin?sectioned, placed on copper grids, and examined under EM, as described in the Materials and Methods section. Prior to the embedding proces, worms are fixed and counterstained with osmium tetroxide. After the worms have been embedded and thin sectioned, the thin sections of tisue are further counterstained with lead citrate and uranyl acetate. To standardize counterstaining conditions of lead citrate and uranyl acetate, we based our first round of standardizations on the recommendation from Dr. Hal. We first tested multiple time points of lead citrate staining 1, 1.5 and 2 min and uranyl acetate time points at 10, 15, 20, and 25 min. There were no significant diferences in electron dense spots betwen control and experimental samples. These results suggested the possibility that the counterstains may interfere with the visualization of heme. We decided to omit counterstaining with lead citrate because it was more likely to precipitate on the grid. To standardize the uranyl acetate counterstaining we incubated the grids for 0, 6, 8, and 10 min time points to determine the minimum counterstaining required to visualize the tisue using EM. After trying to standardize this round of counterstaining, we were stil not seing significant diferences betwen control and experimental samples. We decided to counterstain using only osmium tetroxide to minimize the possibility of counterstaining precipitate. Experiments were conducted with wild-type worms grown to the mid-L4 stage in 80 ?M heme. The results from these two studies were inconclusive. When DAB is oxidized by hydrogen peroxide, it leaves a wel-characterized, easily identifiable dark, electron dense precipitation product. Images from the two identical experiments conducted, revealed litle diference in staining betwen worms from the 84 control and experimental samples. Sections from nine diferent worms in the first experimental group were evaluated. The images taken were of the worm intestine. Present in these samples were storage granules that, because of their electron dense appearance, they were initialy mistaken for possible heme containing vesicles, indicated by asterisks in Figure 13. However, upon examination of other C. elegans EM images, we determined that these vesicles were not heme. These structures are most likely to be gut granules described as having an electron dense appearance because of the various compounds they contain including carbohydrates, lipids and proteins (yolk granules), or concentrated waste products (ww.ormatlas.org). These gut granules were present in both groups of worms incubated with DAB (controls and stained), as wel as in a control group incubated in the absence of DAB. In these experiments there were electron dense spots that caught our atention, but their presence in both control and stained groups led to the conclusion that we could not analyze the electron dense spots as heme staining. These spots varied in number, size and shape. A few sections from the experimental samples had as many as twelve dense spots, while others had only one or two. In the control samples, there was an average of 4-5 electron dense spots. Most of these spots were iregularly shaped and varied in size but sections from both experimental and control samples had an average number of about five iregularly shaped, electron dense spots (Figure 13). 85 Figure 13: Wild-type worms grown at 80 ?M heme, histochemically stained for heme, and examined at the ultrastructural level. Wild-type worms (P 0 ) were grown at 80 ?M heme. Their eggs were harvested by bleaching, and synchronized overnight in M9 buffer. The following morning, the synchronized L1 larvae were inoculated into 80 ?M heme in mCeHR medium. When the larvae reached the mid-L4 stage, they were procesed for EM as described in the Electron Microscopy section of Materials and Methods. The arows depict DAB staining observed in control and experimental samples. Asterisks depict gut granules. This experiment was performed thre times and each micrograph is a representative image. 13A scale bar = 6.67 ?m. 13B scale bar = 2.13 ?m. 13C scale bar = 3.3 ?m. +H 2 O 2 -H 2 O 2 * * * * -H 2 O 2 * 86 C BA 87 These electron microscopy experiments contained two diferent controls. The first control was incubated with DAB only, while the second control was incubated in the absence of DAB. Sections from the first control group contained 2-5 of the iregularly shaped, electron dense spots sen in the stained group of worms. Sections from worms incubated in the absence of DAB were examined and none of these sections had the iregularly shaped electron dense spots observed in the control samples incubated with DAB only (-H 2 O 2 ). This suggested that there was non- specific DAB staining in our first (DAB only) control group. In conclusion of this series of experiments, comparisons betwen stained and control samples revealed areas of the cel with dark precipitate, presumably DAB staining. The precipitate spots were non-uniform in size, shape, and number per worm section (Figure 13) and could not be atributed to lead citrate or uranyl acetate because we did not counterstained the sections with these compounds. Discusion C. elegans is a natural heme auxotroph but relies on exogenously supplied dietary heme for metabolic functions. In the current study, we have established an in situ heme staining protocol for whole animals using C. elegans. While directly staining animal tisues for heme has been conducted for more than 20 years (7,8,54), to our knowledge, the ability to microscopicaly examine heme uptake and localization in an intact, geneticaly tractable animal under physiologic conditions of nutritional heme levels had not been acomplished. Additionaly, we describe a method to stain worms for heme using a modified DAB method that has not been 88 reported previously. Heme staining studies conducted on a heme auxotroph, the catle tick Boophilus microplus, showed that this animal contains a specialized organele caled a ?hemosome? for sequestration of large quantities of heme that this insect ingests (54). However, these studies were not corroborated with histochemical staining of heme in the intact tisue. Engorged ticks were disected, and their midgut tisues, not their whole bodies, were stained for heme. Isolating and staining tisues from whole animals to study heme has been a standard practice for previous staining studies (7,8). Additional studies using digest cels of the tick and fluorescent porphyrin derivatives, of the tick the catle tick showed heme transport in the cel (15). Unlike catle ticks, C. elegans is translucent, making it ideal for cytological heme localization studies using light microscopy. Establishing the methodology described herein took considerable time and efort as DAB, the chromogenic substrate used to localize heme, reacts with endogenous oxygen radicals to produce significant levels of staining in the samples from the control group with pre-incubation with SOD and CAT, we found that heme staining in C. elegans is does dependent (Figure 7). Furthermore, phylogeneticaly diverse nematodes, especialy several species of parasitic nematodes, do not synthesize heme (13). This heme staining technique could provide valuable information on heme uptake, sequestration, and localization in nematodes. Through the standardization proces, we observed that eggs in gravid worms revealed considerable DAB reactive product in the control group. Furthermore, multiple staining experiments using gravid worms revealed no diference in staining 89 betwen gravid embryos of the control and experimental samples. Plausibly, , during during fixation and permeabilization steps, the embryo eggshel was sufficiently permeabilized to permit DAB entry (mw = 214 g/mol) but did not alow sufficient aces to larger molecules, SOD (mw = 32.5 kDa for the dimer) and CAT (mw = 250 kDa for the tetramer). Based on these observations, we decided to focus on synchronized L4 larvae for future studies. We chose the L4 larval stage because this is the developmental stage at which the intestine of the worm is fully developed but embryogenesis has not yet begun, thus providing us with sufficient information regarding heme transport without confounding results due to variations in tisue permeabilization. We also established a connection betwen the concentration of heme in the growth medium and the intensity of the heme staining observed in the worms. With the exception of worms grown at 1.5 ?M heme, the higher the concentration of heme in the medium, the more intensely the worms stained for heme. The largest diference in staining intensities were observed betwen worms grown at 20 ?M and 100 ?M heme as wel as betwen worms grown at 4 ?M and 40 ?M heme. Interestingly, in the second experiment, the group of worms grown in 1.5 ?M heme stained more intensely than expected. Plausibly, under limiting heme conditions, heme uptake must be maximized to sustain normal growth and development. Indeed, unpublished microaray and qRT-PCR results suggest that a significant proportion of heme responsive genes (hrg?s) are highly upregulated at 1.5 ?M? compared to 4 ?M or 20 ?M ?heme. 90 The diferences in staining paterns betwen worms in earlier and worms of later experiments should also be addresed. Worms stained in earlier experiments exhibited uniform staining throughout their bodies. The punctate staining was observed in worms only in later experiments. There are several explanations for punctate staining, each involving the possible ways in which heme is absorbed into the intestinal cels. One possibility is that heme is absorbed by endocytosis. Heme contained within a vesicle formed by endocytosis, could cause the reaction that oxidizes DAB, therein, leaving localized punctate staining. Another explanation for the punctate staining involves a specialized vesicle into which heme is transported after it has been absorbed into the cel. Heme contained within this could also cause punctate staining. There were two paterns of punctate staining: one which showed punctate staining outside the intestinal lumen, presumably concentrated within the intestinal cels, while the second patern was punctate staining that appeared to be outside the intestinal cels. Although it is expected that heme wil be located in both of these places, it is unclear why the punctate staining would remain segregated. Perhaps this is the limitation of using fixed versus live worms. Another diference to note and further explore is the localization of punctate staining. Consistently, there was heavier staining in the anterior half of the intestine that faded gradualy until there was virtualy no detectable staining in the posterior cels. We speculate that the anterior intestinal cels may be the primary location for heme import and uptake in C. elegans. Consistent with this notion, the anterior portion of the intestine secrete digestive enzymes, take up nutrients, and have 91 membrane-bound organeles and vacuoles, while the posterior portions are active in energy storage, e.g., yolk and lipid vacuoles. Thus the diferences we observe with heme staining throughout the intestine may reflect the diference in heme absorption intrinsic to C. elegans (Figure 9). Furthermore, corroborating evidence supporting this data was obtained using Raman confocal microscopy. Diferences in the histochemical heme staining among the mutants, as wel as betwen the mutant and wild-type worms, suggest that these mutants may have diferences in quantitative levels of heme. Each of these mutant strains of worms was selected as a representative strain from each of the 5 complementation groups. Whether these are qualitative diferences in the precise location of the punctate staining cannot be addresed at the light microscopy level. One heme staining consistency betwen the mutant and wild-type worms was that the punctate staining in the mutants was consistently restricted to the first 5 pairs of cels in the anterior region of the intestine. It is also important to note that even though worms from each of the five complementation groups were isolated based on heme resistance, IQ911 and IQ1068 showed opposite levels of heme staining compared to wild-type worms. IQ911 consistently showed les heme staining while IQ1068 showed more intense heme staining than wild-type. Furthermore, IQ938, IQ731, and IQ828 al showed heme staining that was slightly les than wild-type worms. It is currently not known how these animals are capable of surviving in concentrations of heme that are otherwise toxic to wild-type worms, but there are several conceivable explanations. One explanation, that would explain the les intense heme staining observed in IQ911, 92 IQ938, IQ728, and IQ828, is that these worm strains contain mutations in heme transporter machinery, resulting in import of les heme thus preventing the acumulation of toxic concentrations of intracelular heme. While IQ938, IQ731, and IQ828 al had growth rates similar to wild-type, the IQ911 strain consistently grew slower than wild-type by approximately 12 ? 15 h. This observation is consistent with the notion that this strain reveals lower heme staining compared to wild-type worms. Conversely, strain IQ1068 showed slightly increased heme staining as compared to wild-type worms, possible because this mutant can detoxify or sequester heme diferently than wild type. Taken together, future studies wil be needed to addres the biological mechanism for the observed diferences in heme staining intensity in these mutants. Staining paterns and intensities in these mutants wil be further characterized using heme analogs, RNA interference and heme reporter sensor strains. C. elegans reveals a maternal efect for heme that is demonstrated by diferences in growth and development, as wel as in the intensity of histochemical heme staining. The diferences in staining betwen the thre experimental groups of worms were distinct. The likely sources of these diferences are the levels of heme available to the parent and the amount of heme transported to the progeny during embryogenesis. Notably, the progeny with the highest level of heme, that reached the most advanced stage of development, never became gravid. Presumably, progeny from 500 ?M worms had sufficient heme to undergo development to reproductive fitnes, but initiation of embryogenesis must require heme levels greater than what was provided to them by their mothers. While it sems logical that the F 1 generation 93 could be supplied with sufficient heme to undergo embryogenesis, if the P 0 generation was grown at concentrations of heme > 500 ?M, we found that this was not likely to be the case. Heme concentrations ? 800 ?M are toxic to wild-type worms, and the worms growth arest at the L3 stage (13). Future maternal heme experiments could be conducted with mutants to addres whether or not the heme resistant mutations have any efect on the progeny phenotype. Finaly, we provide a framework and preliminary results for future ultrastructural heme localization studies in C. elegans. While there are a few problems that need to be first addresed and optimized prior to continuing EM studies, the implications for studying heme uptake and localization at the EM level using C. elegans, could have a significant impact in heme transport studies. There are numerous advantages to studying heme uptake in C. elegans at the ultrastructural level. First, C. elegans is a natural heme auxotroph that wil have no endogenous heme to confound results. Another advantage is that the entire cel lineage of C. elegans has been mapped, and the location and number of al somatic cels does not vary betwen individual animals. Furthermore, there are thousands of unpublished EM images of C. elegans available on the World Wide Web, at ww.ormiage.org. Many of these images contain labels identifying diferent cels and celular structures, providing an invaluable tool for heme localization in specific cel types. Previously published studies of heme uptake at the electron microscopy level used animals that are heme prototrophs (7,8) and were conducted either under abnormal physiological conditions (7) or with radiolabeled heme (8) Our current protocol for heme staining in C. elegans provides a starting point from which 94 to pursue heme staining studies at the EM level. However, the conditions to observe heme at the ultrastructural level have to be optimized. Microwave fixation and heme staining using DAB produced good results at the light microscopy level. However, it is obvious that 30 % of staining in the control group, even though the staining was extremely light, confounded our results at the EM level. The following steps wil need to be optimized: Establish a microwave fixation protocol that gives high permeabilization of worms without compromising the quality of the anatomical and morphological structure. This wil permit the DAB to penetrate beter at much lower concentration and for shorter incubation times such that worms from the control group give < 5 % visible staining. Levels of tisue preservation were also varied, so it is quite possible that reducing the time of exposure and controlling energy levels of the microwaves wil improve the tisue structure at the ultrastructure level. This standardization wil take a considerable amount of time because at each step the samples need to be examined at the ultrastructural level. Taken together, our studies suggest that individual parameters wil have to be further optimized to adapt our histochemical in situ staining for heme at the light microscopy level to beter preserve tisue and lower non-specific staining at the ultrastructural level for EM. Once these isues have been addresed, electron microscopy experiments wil provide a powerful tool to further advance our knowledge and understanding of heme uptake and localization in not only C. elegans, but also in parasitic worms. 95 REFERENCES 1. Tsiftsoglou, A. S., Tsamadou, A. I., and Papadopoulou, L. C. (2006). Heme as key regulator of major mamalian celular functions: molecular, celular, and pharmacological aspects. Pharmacol Ther, 111: 327-45. 2. Mense, S. M., and Zhang, L. (2006). Heme: a versatile signaling molecule controlling the activities of diverse regulators ranging from transcription factors to MAP kinases. Cel Res, 16: 681-92. 3. Faler, M., Matsunaga, M., Yin, S., Loo, J. A., and Guo, F. (2007). Heme is involved in microRNA procesing. Nat Struct Mol Biol, 14: 23-9. 4. Hamza, I., Chauhan, S., Haset, R., and O'Brian, M. R. (1998). The bacterial ir protein is required for coordination of heme biosynthesis with iron availability. J Biol Chem, 273: 21669-74. 5. Todd, B. L., Stewart, E. V., Burg, J. S., Hughes, A. L., and Espenshade, P. J. (2006). Sterol regulatory element binding proteinis a principal regulator of anaerobic gene expresion in fision yeast. Mol. Cel. Biol., 26: 2817-2831. 6. Kasik, K., and Le, C. C. (2004). Reciprocal regulation of haem biosynthesis and the circadian clock in mamals. Nature, 430: 467-71. 7. Wyllie, J. C., and Kaufman, N. (1982). An electron microscopic study of heme uptake by rat duodenum. Lab Invest, 47: 471-6. 8. Parmley, R. T., Barton, J. C., Conrad, M. E., Austin, R. L., and Holland, R. M. (1981). Ultrastructural cytochemistry and radioautography of hemoglobin- iron absorption. Exp Mol Pathol, 34: 131-44. 96 9. Weisman, Z., and Kornitzer, D. (2004). A family of Candida cel surface haem-binding proteins involved in haemin and haemoglobin-iron utilization. Mol Microbiol, 53: 1209-20. 10. Kabe, Y., Ohmori, M., Shinouchi, K., Tsuboi, Y., Hirao, S., Azuma, M., Watanabe, H., Okura, I., and Handa, H. (2006). Porphyrin acumulation in mitochondria is mediated by 2-oxoglutarate carier. J Biol Chem, 281: 31729- 31735. 11. Quigley, J. G., Yang, Z., Worthington, M. T., Philips, J. D., Sabo, K. M., Sabath, D. E., Berg, C. L., Sasa, S., Wood, B. L., and Abkowitz, J. L. (2004). Identification of a human heme exporter that is esential for erythropoiesis. Cel, 118: 757-66. 12. Krishnamurthy, P. C., Du, G., Fukuda, Y., Sun, D., Sampath, J., Mercer, K. E., Wang, J., Sosa-Pineda, B., Murti, K. G., and Schuetz, J. D. (2006). Identification of a mamalian mitochondrial porphyrin transporter. Nature, 443: 586-9. 13. Rao, A. U., Carta, L. K., Lesuise, E., and Hamza, I. (2005). Lack of heme synthesis in a fre-living eukaryote. Proc Natl Acad Sci U S A, 102: 4270-5. 14. Granick, S., and Levere, R. D. (1965). The intracelular localization of heme by a fluorescence technique J. Cel Biol., 26: 167-176. 15. Lara, F. A., Lins, U., Bechara, G. H., and Oliveira, P. L. (2005). Tracing heme in a living cel: hemoglobin degradation and heme trafic in digest cels of the catle tick Boophilus microplus. J Exp Biol, 208: 3093-101. 16. Liem, H. H., Tavasoli M., Muller-Eberhard, U. (1975). Celular and 97 subcelular localization of heme and hemopexin in the rabbit. Acta Haematology, 53: 219-225. 17. Graham, R. C., Jr., Karnovsky, MJ (1966). The early stages of absorption of injected horseradish peroxidase in the proximal tubules of mouse kidney: ultrastructural cytochemistry by a new technique. J. Histochem. Cytochem., 14: 291-302. 18. Ajioka, R. S., Philips, J. D., and Kushner, J. P. (2006). Biosynthesis of heme in mamals. Biochim Biophys Acta, 1763: 723-36. 19. Baros, M. H., Nobrega, F. G., and Tzagoloff, A. (2002). Mitochondrial feredoxin is required for heme A synthesis in Sacharomyces cerevisiae. J Biol Chem, 277: 9997-10002. 20. Morrison, M. S., Crico, J.A., Hegg, E.L. (2005). The biosynthesis of heme O and heme A is not regulated by copper. Biochemistry, 44: 12554-12563. 21. Moraes, C. T., Diaz, F., and Barientos, A. (2004). Defects in the biosynthesis of mitochondrial heme c and heme a in yeast and mamals. Biochim Biophys Acta, 1659: 153-9. 22. Li, J. M., Brathwaite, O., Cosloy, S.D., Russel, C.S (1989). 5-aminolevulinic acid synthesis in escherichia coli. Journal of Bacteriology, 171: 2547-2552. 23. Panek, H., and O'Brian, M. R. (2002). A whole genome view of prokaryotic haem biosynthesis. Microbiology, 148: 2273-82. 24. Nakajima, O., Takahashi, S., Harigae, H., Furuyama, K., Hayashi, N., Sasa, S., and Yamamoto, M. (1999). Heme deficiency in erythroid lineage causes diferentiation arest and cytoplasmic iron overload. Embo J, 18: 6282-9. 98 25. Dioum, E. M., Rutter, J., Tuckerman, J. R., Gonzalez, G., Giles-Gonzalez, M.-A., and McKnight, S. L. (2002). NPAS2: A gas-responsive transcription factor. Science, 298: 2385-2387. 26. Giles-Gonzalez, M.-A., and Gonzalez, G. (2005). Heme-based sensors: defining characteristics, recent developments, and regulatory hypotheses. Journal of Inorganic Biochemistry, 99: 1-22. 27. Han, L., Lu, J., Pan, L., Wang, X., Shao, Y., Han, S., and Huang, B. (2006). Histone acetyltransferase p300 regulates the transcription of human erythroid- specific 5-aminolevulinate synthase gene. Biochemical and Biophysical Research Communications, 348: 799-806. 28. Melefors, O., Goossen, B., Johansson, H.E., Stripecke, R., Gray, N., Hentze, M.W. (1993). Translational control of 5-aminolevulinate synthase mRNA by iron-responsive elements in erythroid cels. The Journal of Biological Chemistry, 268: 5974-5978. 29. Cooperman, S. S., Meyron-Holtz, E. G., Oliviere-Wilson, H., Ghosh, M. C., McConnel, J. P., and Rouault, T. A. (2005). Microcytic anemia, erythropoietic protoporphyria, and neurodegeneration in mice with targeted deletion of iron-regulatory protein 2. Blood, 106: 1084-1091. 30. Dwyer, B. E., Stone, M. L., Zhu, X., Pery, G., and Smith, M. A. (2006). Heme deficiency in Alzheimer's disease: a possible connection to porphyria. J Biomed Biotechnol, 2006: 24038. 31. Oates, P. S., and West, A. R. (2006). Heme in intestinal epithelial cel turnover, diferentiation, detoxification, inflamation, carcinogenesis, 99 absorption and motility. World J Gastroenterol, 12: 4281-95. 32. Hughes, A. L., Powel, D. W., Bard, M., Eckstein, J., Barbuch, R., Link, A. J., and Espenshade, P. J. (2007). Dap1/PGRMC1 binds and regulates cytochrome P450 enzymes. Cel Metabolism, 5: 143-149. 33. Fayadat, L., Nicoli-Sire, P., Lanet, J., and Franc, J.-L. (1999). Role of heme in intracelular traficking of thyroperoxidase and Involvement of H2O2 generated at the apical surface of thyroid cels in autocatalytic covalent heme binding. J. Biol. Chem., 274: 10533-10538. 34. Ye, W., and Zhang, L. (2004). Heme controls the expresion of cel cycle regulators and cel growth in HeLa cels. Biochem Biophys Res Commun, 315: 546-54. 35. Maywood, E. S., Mrosovsky, N., Field, M. D., and Hastings, M. H. (1999). Rapid down-regulation of mamalian Period genes during behavioral reseting of the circadian clock. PNAS, 96: 15211-15216. 36. Davies, B. S. J., and Rine, J. (2006). A role for sterol levels in oxygen sensing in sacharomyces cerevisiae. Genetics, 174: 191-201. 37. Shan, Y., Lambrecht, R. W., Ghaziani, T., Donohue, S. E., and Bonkovsky, H. L. (2004). Role of Bach-1 in regulation of heme oxygenase-1 in human liver cels: insights from studies with smal interfering RNAS. J Biol Chem, 279: 51769-74. 38. Genco, C. A., and Dixon, D. W. (2001). Emerging strategies in microbial haem capture. Mol Microbiol, 39: 1-11. 39. Latunde-Dada, G. O., Simpson, R. J., and McKie, A. T. (2006). Recent 100 advances in mamalian haem transport. Trends Biochem Sci, 31: 182-8. 40. Wandersman, C., and Delepelaire, P. (2004). BACTERIAL IRON SOURCES: From Siderophores to Hemophores. Annual Review of Microbiology, 58: 611-647. 41. Skaar, E. P., Humayun, M., Bae, T., DeBord, K. L., and Schneewind, O. (2004). Iron-source preference of Staphylococcus aureus infections. Science, 305: 1626-8. 42. Brickman, T. J., Vanderpool, C. K., and Armstrong, S. K. (2006). Heme transport contributes to in vivo fitnes of Bordetela pertussis during primary infection in mice. Infect Imun, 74: 1741-4. 43. Wyckoff, E. E., Lopreato, G. F., Tipton, K. A., and Payne, S. M. (2005). Shigela dysenteriae ShuS promotes utilization of heme as an iron source and protects against heme toxicity. J. Bacteriol., 187: 5658-5664. 44. Santos, R., Buison, N., Knight, S., Dancis, A., Camadro, J. M., and Lesuise, E. (2003). Haemin uptake and use as an iron source by Candida albicans: role of CaHMX1-encoded haem oxygenase. Microbiology, 149: 579-88. 45. Pendrak, M. L., Chao, M. P., Yan, S. S., and Roberts, D. D. (2004). Heme oxygenase in Candida albicans is regulated by hemoglobin and is necesary for metabolism of exogenous heme and hemoglobin to alpha-biliverdin. J Biol Chem, 279: 3426-33. 46. Colley, D. G., LoVerde, P. T., and Savioli, L. (2001). Infectious disease. Medical helminthology in the 21st century. Science, 293: 1437-1438. 47. Sangster, N. C., and Gil, J. (1999). Pharmacology of anthelmintic resistance. 101 Parasitol Today, 15: 141-6. 48. Held, M. R., Bungiro, R. D., Harison, L. M., Hamza, I., and Cappelo, M. (2006). Dietary iron content mediates hookworm pathogenesis in vivo. Infect Imun, 74: 289-95. 49. Foster, J., Ganatra, M., Kamal, I., Ware, J., Makarova, K., Ivanova, N., Bhatacharyya, A., Kapatral, V., Kumar, S., Posfai, J., Vincze, T., Ingram, J., Moran, L., Lapidus, A., Omelchenko, M., Kyrpides, N., Ghedin, E., Wang, S., Goltsman, E., Joukov, V., Ostrovskaya, O., Tsukerman, K., Mazur, M., Comb, D., Koonin, E., and Slatko, B. (2005). The Wolbachia genome of Brugia malayi: endosymbiont evolution within a human pathogenic nematode. PLoS Biol, 3: e121. 50. Krishnamurthy, P., and Schuetz, J. D. (2005). The ABC transporter Abcg2/Bcrp: role in hypoxia mediated survival. Biometals, 18: 349-58. 51. Jonker, J. W., Buitelar, M., Wagenaar, E., van der Valk, M. A., Schefer, G. L., Scheper, R. J., Plosch, T., Kuipers, F., Elferink, R. P. J. O., Rosing, H., Beijnen, J. H., and Schinkel, A. H. (2002). The breast cancer resistance protein protects against a major chlorophyll-derived dietary phototoxin and protoporphyria. PNAS, 99: 15649-15654. 52. Hamza, I. (2006). Intracelular traficking of porphyrins. ACS Chem Biol, 1: 627-9. 53. Qiu, A., Jansen, M., Sakaris, A., Min, S. H., Chatopadhyay, S., Tsai, E., Sandoval, C., Zhao, R., Akabas, M. H., and Goldman, I. D. (2006). Identification of an intestinal folate transporter and the molecular basis for 102 hereditary folate malabsorption. Cel, 127: 917-28. 54. Lara, F. A., Lins, U., Paiva-Silva, G., Almeida, I. C., Braga, C. M., Miguens, F. C., Oliveira, P. L., and Dansa-Petretski, M. (2003). A new intracelular pathway of haem detoxification in the midgut of the catle tick Boophilus microplus: aggregation inside a specialized organele, the hemosome. J Exp Biol, 206: 1707-1715. 55. Xie, G., Jia, Y., and Aamodt, E. (1995). A C. elegans mutant scren based on antibody or histochemical staining. Genet Anal, 12: 95-100. 56. Sciaco, M., and Bonila, E. (1996). Cytochemistry and imunocytochemistry of mitochondria in tisue sections. Methods Enzymol, 264: 509-21. 57. Paupard, M. C., Miler, A., Grant, B., Hirsh, D., and Hal, D. H. (2001). Imuno-EM localization of GFP-tagged yolk proteins in C. elegans using microwave fixation. J Histochem Cytochem, 49: 949-56. 58. Brenner, S. (1974). The genetics of Caenorhabditis elegans. Genetics, 77: 71- 94. 59. Grad, L. I., and Lemire, B. D. (2004). Mitochondrial complex I mutations in Caenorhabditis elegans produce cytochrome c oxidase deficiency, oxidative stres and vitamin-responsive lactic acidosis. Hum. Mol. Genet., 13: 303-314. 60. Simionescu, N., Simionescu, M., and Palade, G. E. (1973). Permeability of muscle capilaries to exogenous myoglobin J. Cel Biol., 57: 424-452. 61. Kachur, T., Ao, W., Berger, J., and Pilgrim, D. (2004). Maternal UNC-45 is involved in cytokinesis and colocalizes with non-muscle myosin in the early Caenorhabditis elegans embryo. J Cel Sci, 117: 5313-5321. 103 62. Kamping, A., Katju, V., Beukeboom, L. W., and Weren, J. H. (2006). Inheritance of gynandromorphism in the parasitic wasp Nasonia vitripennis. Genetics: genetics.106.067082. 63. Garamszegi, L. Z., Biard, C., Eens, M., Moller, A. P., Saino, N., and Surai, P. Maternal efects and the evolution of brain size in birds: Overlooked developmental constraints. Neuroscience & Biobehavioral Reviews, In Pres, Corrected Proof. 64. Giordano, M., and Momigliano-Richiardi, P. (2004). Maternal efect in multiple sclerosis. The Lancet, 363: 1748-1749. 65. Fukushige, T., Goszczynski, B., Yan, J., and McGhee, J. D. (2005). Transcriptional control and paterning of the pho-1 gene, an esential acid phosphatase expresed in the C. elegans intestine. Developmental Biology, 279: 446-461. 66. Burges, J., Hihi, A. K., Benard, C. Y., Branicky, R., and Hekimi, S. (2003). Molecular mechanism of maternal rescue in the clk-1 mutants of Caenorhabditis elegans. J. Biol. Chem., 278: 49555-49562.