HM Medical Clinic

 

Microsoft word - thesis 1.3.doc

The Extraction of Native Giardia lamblia Actin
Through Differential Detergent Treatment
Nicholas Barker Stoler Submitted in fulfillment of the Senior Thesis requirement in the Department of Biology, Georgetown University, Washington, D.C., May 2008 NICHOLAS B. STOLER The Extraction of Native Giardia lamblia Actin Through Differential Detergent
Treatment
Dr. Heidi G. Elmendorf, Department of Biology, Georgetown University Giardia lamblia is an intestinal pathogen which is universally encountered in the third world. Giardia is a protozoan parasite which must attach to the intestinal wall in order to remain in the digestive tract and thrive. It attaches with its ventral surface to the intestinal endothelium. Previous research by our lab has suggested that actin-based microfilaments direct this process. Inhibiting the function of actin would thus be very effective at reducing disease pathology. Studies to better understand Giardia actin are most effective with a high-fidelity version of the protein. The Giardial version of actin is rare in that it is only 60% identical to human actin, while actin is well-conserved amongst most other eukaryotes. The simplest expression system to produce this actin exogenously would use E. coli, but that may result in a non-native conformation due to the lack of eukaryotic folding factors. Also, a standard eukaryotic expression system using Baculovirus has been attempted by our lab but resulted in poor yields. Thus in order to produce the most accurate version of the protein, I have refined procedures towards extracting actin from Giardia cells themselves. The focus of this study was a strategy for detergent-extracting Giardia cells so that cellular contents are removed before dissociating an actin-enriched eluate. I have developed a preliminary detergent buffer and methods of visualizing the extraction process. Following extraction, I will then further purify the actin in the eluate. The actin produced by this method will be useful for drug-binding studies, actin polymerization kinetics studies, electron microscopic investigation of microfilament morphology, and the search for actin-associated proteins in Giardia. I would like to thank everyone who has supported me and helped me through my entire tenure at the Elmendorf lab. First, Dr. Heidi Elmendorf, who has created an atmosphere that is both relaxed and conscientiously inquisitive. Her scientific nurturing through the years has equipped me with the most useful tools I could ask for in the real world of science. I also deeply appreciate the advocacy and generosity of Dr. Joseph Neale and the rest of the Howard Hughes Program at Georgetown. I would have far fewer years of experience behind me if it weren't for them. Next, I would like to thank the many members of the Elmendorf lab, especially Jesse Cohen and Colleen Walls, who over the years have helped me learn everything I know about working in a lab. Finally I want to express my gratitude to my friends and especially my family for enduring all my concerns about my tasks and supporting me through it all. Table of Contents
Chapter I. Introduction A. Giardia and giardiasis B. The Giardia Cytoskeleton C. Mechanism of Attachment D. Targeting Actin E. Isolation of Giardia Actin Chapter II. Materials and Methods A. Culturing Giardia lamblia B. Preparation of Giardia for fluorescent staining and microscopic C. Fluorescent labeling of Giardia microfilaments with Cytochalasin D-BODIPY FL D. Fluorescent labeling of Giardia membranes with CM-DiI E. Fluorescent labeling of Giardia membranes with BODIPY FL C5-ceramide F. Light microscope observation of detergent treatment of Giardia lamblia Chapter III. Results A. Microscopic observations of detergent treatment of Giardia B. Fluorescent labeling of Giardia microfilaments with Cytochalasin D-BODIPY FL C. Fluorescent labeling of Giardia outer membranes with CM-DiI D. Fluorescent labeling of Giardia membranes with BODIPY FL C5-ceramide Chapter IV. Discussion Giardia and giardiasis
Giardia lamblia is a worldwide health problem that requires better solutions. Giardia lamblia is the causative agent of giardiasis, a disease which affected over 200 million people per year in 2002 (Lane and Lloyd, 2002). giardiasis is a major diarrheal disease worldwide, and diarrheal diseases account for a quarter of deaths under the age of 5 (Bryce et al., 2005). Side effects of and growing resistance to current treatments is a problem, giving importance to the search for new drugs. The Giardia lamblia pathogen itself is a potentially early diverging unicellular protozoan. It is of the family Hexamitidae in the order Diplomonadida, indicating its twin nuclei. It is thought to be of a lineage which diverged soon after the evolution of eukaryotes. One reason for this hypothesis is the species' lack of typical eukaryotic features such as golgi apparati and mitochondria. However, recently evidence has been found for mitochondrial and golgi-related genes. This indicates the presence of these organelles at some point in its evolutionary history, meaning Giardia did not diverge before their existence (Dacks et al., 2003; Emelyanov, 2003). But evidence remains in rRNA sequence phylogenetics that G. lamblia is a basal eukaryote (Sogin et al., 1989). A consequence of this distinction is that many giardial systems are very divergent compared to their human and eukaryotic counterparts. When symptomatic, Giardia infection causes malabsorptive diarrhea that poses a major problem for most of the world. Most infections are asymptomatic, but when they present symptoms they usually produce persistent diarrhea lasting over two weeks (Ortega and Adam, 1997). This can be life-threatening in the third-world, especially in children. Between 2000 and 2003, diarrheal diseases caused 18% of all deaths in children younger than 5, lower than only pneumonia (Bryce et al., 2005). In a study at a hospital in Mozambique in 2001, 2.5% of children under 5 admitted for diarrhea were infected with Giardia (Mandomando et al., 2007). Even when the infection is cleared, the effect on childhood development remains. Affected children have more trouble growing and gaining weight compared to those spared of infection (Fraser et al., 2000). The effect on livestock is also significant, with a Canadian study showing 100% prevalence of Giardia in beef calves (Ralston et al., 2003). Infection inhibits livestock weight gain while simultaneously increasing required feed usage (Olson et al., 2004). Usually-effective drug treatments exist for giardiasis, but there are problems with the current repertoire including side effects and recurrence. The current standard is metronidazole, with alternatives being tinidazole, furazolidone, and quinacrine. Difficulties include the fact that furazolidone can cause complications in patients of a certain genotype and quinacrine causes significant side effects. A major issue with current drugs is the issue of recurrence. After metronidazole treatment, the persistence of chronic giardiasis can be as high as 32% in the following 7 months (Hanevik et al., Resistance is also an increasing problem for many of these drugs. Resistance to metronidazole is up to 20% as of 2007. Often, these drugs will inhibit the same cellular process in Giardia. Thus, resistance to one will also act as resistance to another, such as the case of metronidazole and tinidazole (Ali and Nozaki, 2007). To truly address the problem of resistance, new molecular targets for drugs have to be found. The pathology of giardiasis is caused by intestinal colonization by the trophozoite form of the parasite. Giardia alternates between the trophozoite and a cyst stage. The trophozoite form is more metabolically active while the cyst is seen as mostly dormant. Trophozoites will colonize the duodenum and jejunum and absorb nutrients in order to grow and multiply (Astiazaran-Garcia et al., 2000). The covering of the intestinal absorptive surface by Giardia may cause the malabsorption, as may the morphological changes they induce in intestinal villi (Adam, 1991). Some trophozoites are then forced further down the small intestine and differentiate into the cyst form (Luján et al., 1996). These cysts are eliminated with the rest of the gastrointestinal waste and can survive harsh environmental conditions (Gillin et al., 1996). When cysts are ingested, they respond to low gastric pH and then higher intestinal pH by excysting back into the trophozoite form (Boucher and Gillin, 1990). The Giardia Cytoskeleton
An understanding of the eukaryotic cytoskeleton is essential to understanding the source of a cell's shape and structural strength, as well its motility. The Giardia cytoskeleton is important in supplying each of these characteristics, with the addition of a fourth: pathogenicity. Protozoan cytoskeletons are typically composed of two different types of filaments: microtubules and microfilaments. Only microtubules and microfilaments have been described in Giardia (Elmendorf et al., 2003). Microtubules are composed of the proteins α- and β-tubulin polymerized into strong tubes 25 nm in diameter. Microtubules are the structural components of flagella and their basal bodies. Microfilaments are narrower (5-9 nm) fibres composed of polymerized actin monomers. The monomers stack one on top of the other two proteins wide and progress in a helix. In most eukaryotic cells, numerous different proteins bind to tubulin and actin their polymerized and monomeric forms. While a number of microtubule-associated proteins are known in Giardia (Shapiro, 2006), currently no actin-binding proteins have been discovered (Elmendorf et al., 2003). The Giardia cytoskeleton contains several ultrastructural features discernable by light microscopy. To locate these features, the basic anatomy of the trophozoite must be understood. They are bilaterally symmetrical teardrop-shaped cells with distinct dorsal, ventral, anterior, and posterior ends. Features include a ventral disk crucial to attachment, a microtubule-based median body central in the cell, a funis of microtubules stretching between the caudal and posterior axonemes, and four pairs of flagella (Elmendorf et al., 2003). Each of these structures is composed of microtubules a main structural element, along with many other associated proteins. Of interest to this study, there is often evidence of actin being one of the associated proteins of a particular These cytoskeletal elements are largely interconnected. The microtubules which form the ventral disk can be seen to emerge from the flagellar basal bodies in a spiral. Also, the axonemes of the posterior flagella are linked along their length to the caudal axonemes by fibers of the funis (Benchimol et al., 2004). Even the median body seems to be anchored somewhat to the dorsal plasma membrane (Piva and Benchimol, 2004). The nuclei too are linked to the cytoskeleton through the axonemes (Benchimol, 2005). The median body is a largely uncharacterized microtubule-based structure which lies posterior to the nuclei. It appears as a bundle of rods slightly curved in shape which almost spans the width of the cell. It consists of microtubules bundled in no clear pattern. This grouping of microtubules leads some to suggest it functions as a store of prefabricated microtubules (Piva and Benchimol, 2004). Other proteins which are associated are actin, α-actinin, β-giardin and a coiled-coil protein specific to the median body (Feely et al., 1982; Marshall and Holberton, 1993). Unlike most structures in Giardia, the flagella appear to be similar those of most other protozoa. They exhibit the standard 9+2 microtubule arrangement, and originate in basal bodies near the midpoint of the nuclei (Friend, 1966). There are eight total flagella, arranged in four pairs which extend from various points on the cell surface. The anterior pair run from their basal bodies toward the anterior end of the cell, then curve back around to emerge from the anterior end in the caudal direction. The ventral flagella surface on the ventral side of the parasite under the ventral disk. They lie in a groove in the ventral surface, running posteriorly. The caudal flagella run straight from the basal bodies down the centerline of the cell, then out the caudal tip of the tail. The posterior- lateral flagella run from the basal bodies out toward the posterior sides of the cell body. They form an acute angle with the caudal flagella so that the distance between the two pairs of flagella increases as they run toward the back of the cell. Eventually the posterior-lateral flagella emerge from the posterior sides of the cell (Elmendorf et al., 2003; Benchimol et al., 2004). Many intracellular portions of the axonemes of the flagella are accompanied by other structures as they run through the cytoplasm. Giardia's flagella are unusual in that their axonemes often have long intracellular portions. Thus the structures associated with these extended intracellular axonemes are not found in other organisms. A series of electron-dense material called the "striated fibres" runs alongside the anterior flagella and also contacts the ventrolateral flange (Holberton, 1973). "Dense rods," another type of electron-dense material, have also been described running alongside the anterior and posterior-lateral axonemes. These dense rods have shown localization of centrin (Correa et al., 2004). Also, actin has been found along the basal bodies and near the dense rods of the posterior-lateral axonemes (Feely et al., 1982; Narcisi et al., 1994). Then along the caudal flagella runs the most complex electron-dense material, the funis. The funis is a complex structure located in the caudal complex or "tail" of the cell. It begins as two sheets of microtubules which run along the caudal axonemes. The sheets of microtubules are on the ventral side of one caudal axoneme and on the dorsal side of the other. Each sheet then tapers off as its constituent microtubules veer off laterally one at a time towards one of the posterior-lateral axonemes. The funis microtubules attach to the posterior-lateral axonemes, forming a net covering the area between those axonemes and the caudal ones (Benchimol et al., 2004). Benchimol et al. further propose that the funis microtubules attach specifically to the dense rods running along the posterior-lateral axonemes, though this remains unclear. Parallel funis microtubules in this net are themselves linked to each other by filamentous bridges. After the posterior-lateral axonemes exit the cell body, the microtubules are anchored to a filamentous network associated with the plasma membrane. The microtubules have been observed covered by an unidentified electron-dense material (Benchimol et al., 2004). The ventral disk structure dominates the entire Giardia ultrastructure. The disk is a domed structure with a space in the center which sits just under the ventral plasma membrane. The structure is based on a spiral whorl of microtubules extending from the gap in the center. The part which overlays the caudal axonemes is reduced in radius, resulting in a notch cut into the disk in that location (Benchimol et al., 2004). Another component of the ventral disk is the microribbons. These structures extend anteriorly from each microtubule towards the cell interior. The microribbons are composed mostly of proteins called giardins (Peattie et al., 1989). β-giardin is a structural protein which spontaneously forms 2.5 nm fibers (Crossley and Holberton, 1985), predicted to form an extended coiled-coil structure (Holberton et al., 1988). Adjacent microribbons are connected to each other by linear structures termed crossbridges (Peattie et al., 1989; Holberton, 1973). The crossbridges aid in holding together the whole disk structure (Holberton, 1981). As the disk continues towards the periphery of the cell, it tapers off into the lateral crest, a denser network of fibers. Actin has been localized to this periphery of the disk, possibly aiding in contraction (Feely et al., 1982). Further beyond the end of the disk, the cell body extends into a flap called the ventrolateral flange. This flap can protrude from just above the disk down towards the attachment substrate below (Elmendorf et al., 2003). Mechanism of Attachment
In order to survive the turbulent conditions in the small intestine, Giardia cells must attach to the intestinal wall using an unresolved mechanism. Peristaltic movements within the small intestine create fluid flows which will expel any contents not secured to the intestinal wall. In order for Giardia cells to remain in their nutrient-rich habitat and multiply, they must have a strategy to remain in place. At least four different theories compete to explain Giardia's attachment in different ways. Common themes run through many of the four theories. The first relies on lectin binding, a usual method of cell-cell adhesion. The other three focus on the involvement of the cytoskeleton. The first proposes that Giardia "clutch" the substrate with the lateral crest of the adhesive disk in a grasping motion. The remaining two rely on the concept of a suction force between the Giardia cell and its substrate. One proposes that fluid flow caused by the beating of flagella dynamically creates a region of lower, or "negative" pressure under the cell. The other proposes that this negative pressure is formed by a "suction cup"-type contraction of the ventral disk. Recent focus has been on the latter three models which propose a mechanical, not biochemical, basis of attachment. Opinion currently disfavors the theory that surface lectin binding is the major mediator in attachment. Some studies have suggested this theory in the past, showing that inhibiting lectins with free sugars reduced Giardia attachment to intestinal cells (Inge et al., 1988). But further studies have disputed this purported effect (Magne et al., 1991) and the theory has trouble explaining the ease of attachment to hydrophobic substrates, glass, and plastic (Hansen et al., 2006) as well as the extreme preference for attaching with the ventral surface. The true role of lectins seems to be to augment that of mechanical mechanisms (Sousa et al., 2001) An alternative theory is that the lateral crest engages in a "clutching" action, holding parasite to enterocyte. Epithelial cells are often damaged after attachment, left with an indentation in the shape of the lateral crest (Erlandsen and Chase, 1974). This suggests a mechanism where the lateral crest somehow intercalates itself into the microvillous brush border of enterocytes. This could also be achieved by contracting the ventral disk. The presence of contractile proteins in the periphery of the ventral disk supports this theory (Feely et al., 1982). Another study, by Erlandsen et al. (2004) suggests that the ventrolateral flange possesses some adhesive activity which helps to initiate the clutching attachment by the lateral crest. However, to be clear, the initial help which they propose the ventrolateral flange gives the cell in correctly orienting itself could also act in conjunction with any of the other attachment hypotheses. Another mechanical model, of flagella creating negative pressure through fluid flow, is problematic. The idea, proposed by Holberton in 1974, was prompted by the observation that the flagella constantly oscillate while the cell is attached. The idea is that Giardia's beating flagella directing a flow of fluid under the parasite, in-between the ventrolateral flange and the lateral crest. Through a connection to the chamber under the ventral disk, this fluid flow would create a zone of lowered pressure under the parasite. This negative-pressure zone directly between the parasite and intestinal cell would draw the two together. One issue with this is that the ventral flagella emerge in a different place than was thought at the time of Holberton (Erlandsen and Feely, 1984), changing the calculated fluid dynamics. Because of this issue and recent findings, favor has shifted to another negative-pressure hypothesis. This hypothesis, the "suction cup" theory, is based on the idea of the ventral disk contracting upon contact, thus pulling up on the chamber between the cells and generating a negative pressure. It is possible that the ventral disk itself is capable of contracting, decreasing its diameter and thus becoming more concave (Sousa et al., 2001). Evidence for contractile proteins in the periphery of the disk supports this hypothesis (Feely et al., 1982). This increasing concavity of the disk would literally act as a suction cup: increasing the volume sealed between the disk and enterocyte would lower the fluid pressure. This decreased (or "negative") pressure would create a force holding together the surrounding two cells. It can also be imagined how a strong enough suction force could leave behind the marks seen on enterocytes in Erlandsen and Chase (1974). Also, unpublished data collected by our lab unambiguously shows cells attached to a glass surface and moving across the surface at the same time. A negative pressure model is the only one which could conceivably fit such results. Still, the study by Erlandsen et al. (2004) shows cells attaching in situations where the possibility of forming negative pressure has been eliminated. In this study, the authors created a polystyrene attachment substrate composed of raised pillars less wide than the ventral disk. The gaps between these pillars created an inevitable leak in the negative pressure cavity. Some cells still attached to the pillars, showing that the generation of negative pressure is not the only non-lectin mechanism of attachment. But the creation of pillars reduced attachment rates more than sixfold, indicating the overall dominance of the negative pressure mechanism. Targeting Actin
A number of studies have shown the centrality of the cytoskeleton in attachment, and the more tentative predominance of actin over tubulin. Many studies have treated parasites with microtubule or microfilament-disrupting drugs, and seen inhibition of attachment. The results have been very inconsistent, which may be the result of a number of different drug concentrations, attachment substrates, and treatment protocols. But generally microfilament-disrupting drugs have shown greater inhibition of attachment at lower concentrations than microtubule-affecting drugs (Elmendorf et al., The importance of microfilaments in attachment combined with its highly divergent sequence highlights it as a potential drug target. In order to serve as an effective target, however, Giardia actin must differ structurally from mammalian (human) actin. Drugs which bind to and inhibit Giardia actin will detach the parasite but will be ineffective if the same drug kills the adjacent host cell next to it by binding its actin. Sufficient divergence would be normally unexpected in actin. It is usually very strictly conserved, averaging 80-85% sequence identity in fungi, plants, and metazoa (Doolittle and York, 2002). But fortunately Giardia actin diverges greatly, on average only 58% identical to other species' actin (Drouin et al., 1995). Tubulin, however, is more conserved than actin (Adam 2001). A quick NCBI BLAST search shows Giardia β-tubulin to be about 88% identical to the human version (accession numbers XP_001707372 and NP_821080, respectively). The highly divergent nature of Giardia actin is made very clear by the failure of standard microfilament drugs to take effect. Phalloidin-conjugated probes have been unable to stain Giardia microfilaments (Elmendorf, unpublished data). Jasplakinolide, which binds to the same surface as phalloidin, is similarly unable to inhibit attachment at any concentration. If attachment is actin-mediated, then this gives further evidence that Giardia actin is jasplakinolide-insensitive. It still produces morphological disruptions, though, at high concentrations (Carvalho and Monteiro-Leal, 2004). Cytochalasin B is a known microfilament inhibitor, and it has shown its potential to affect Giardia. It has been shown to disrupt cytoskeletal structure by Correa and Benchimol (2006). It has also been shown to disrupt attachment, but not unambiguously (Roskens and Erlandsen, 2002; Sousa et al., 2001). Isolation of Giardia Actin
In order to assay Giardia actin for its binding properties, it must be isolated and purified. The gene has been cloned, and so can be expressed simply with a standard prokaryotic expression system (Yin et al., 2007). However, the eukaryotic actin gene will likely not fold properly in a prokaryote. Actin folding is known to require chaperones like CCT and prefoldin in yeast, otherwise emerging with a non-native tertiary structure (Vainberg et al., 1998; Speiss et al., 2004). This could cause it to have different binding and polymerization properties than Giardia actin in vivo as well as different filament morphologies. Thus a eukaryotic expression system is desirable. A possible eukaryotic method of expression is the baculovirus system, though this has not yielded great progress so far in our lab. Previous work in an Apicomplexan protist, Toxoplasma gondii, showed success in using a baculovirus expression system to obtain actin. This system uses a baculovirus to transfect the desired actin gene into insect cells, a eukaryotic cell line. The insect cells then express this actin gene, which has been tagged with a 6xHis moiety consisting of six consecutive histidine residues. This arrangement of residues chelates nickel atoms with high affinity, allowing the use of nickel-agarose affinity chromatography to purify the extracted actin (Sahoo et al., 2006). This avenue is currently being pursued in our lab, but it so far had difficulty producing actin in quantity. In any case, it still is an exogenous expression system with the possibility of yielding a protein with non-native conformation. Thus a strategy to produce the highest-fidelity actin is to extract actin directly from Giardia itself. Similar work has again been done in T. gondii to extract its actin. Preceding studies had used high-resolution Field Emission Scanning Electron Microscopy (FESEM) to visualize actin filaments associated with the cytoskeletal subpellicular network of T. gondii (Schatten et al., 2003). Patrón et al. then isolated this structure using a detergent extraction to obtain the actin associated with it. They purified the actin from the subpellicular network with DNase I affinity chromatography (Patrón et al., 2005). There is similar FESEM work which has been done in Giardia. Benchimol et al. have produced cytoskeletal preparations which seem to have preserved some microfilament material. In some FESEM images there is a visible amount of membrane-associated fibrillar material in the caudal complex underneath the area where the funis branches out to contact the posterior-lateral axonemes. This fibrillar material may contain membrane-associated actin. Other areas which seem to contain this fibrillar material have been identified as structures where actin localizes. Such areas are the basal bodies and the dense rods associated with the posterior-lateral axonemes (Feely et al., 1982; Narcisi et al., 1994; Benchimol et al., 2004; Sant'Anna et al., 2005). Still, the appearance of the material is ambiguous. But the elucidation of its nature is aided by a number of reasons why these areas should contain microfilaments (Benchimol, 2005). First, actin typically nucleates at the plasma membrane and forms part of the cortical cytoskeleton underlying the plasma membrane (Welch and Mullins, 2002). Also, microtubules are clearly not present in any of the preparations, and the shape of the dorsal membrane suggests that it requires cytoskeletal support. Thus, microfilaments are likely to fill that role instead (Elmendorf et al., 2003). Also, the movement of the caudal complex during motility suggests a role there for contractile structures like microfilaments (Benchimol et al., 2004). The same studies by Benchimol et al. show the promise of a detergent treatment to specifically dissociate actin. In the same studies showing cytoskeleton preparations preserving the actin-like material, very similar preparations are shown where the actin material has dissociated. The preparation seems to have used a very similar procedure, implying that a small change in conditions may cause actin to be released from the cells with minimal contamination (Benchimol, 2005). The material could then be isolated from the detergent-treated cells, yielding a partially purified actin sample. However, elucidating the best treatment to dissociate actin is difficult using only the published work by Benchimol et al. It is not clear what precise treatment yields the preparations shown with actin and the preparations shown without actin. The most important step, the method of proceeding from the former preparation to the latter, is especially unclear. The technique could call for changing buffers, a physical step such as agitation, or simply letting the sample incubate longer. Thus if I am to use this information, I will have to complete the protocol myself through experimentation. My specific procedure will be based on the above studies, adjusted for my microfilament-specific goals. I will attempt to use a similar detergent extraction procedure to Benchimol et al. to remove most of the cell body with one step of detergent extraction, then further treat with detergent to obtain a fraction enriched with actin. The buffers will be based on the PHEM buffer (PIPES, HEPES, EGTA, MgCl2) used by Benchimol et al.. This is a common cytoskeletal preparation buffer which is used to stabilize microtubules but also appears to do the same for microfilaments (Bell et al., 1989). In the buffer to be used to extract the cell bulk while preserving microfilaments, I have added ATP, which is necessary to maintain actin polymerization (Welch and Mullins, 2002). I have included ATP because strategy will begin with the goal of preserving actin through the first phase of extraction. Then to cause actin to maximally dissociate, I may have to add a different buffer without ATP and possibly Mg2+, which also helps maintain polymerization. Thus I will refine my procedure over time based on In order to obtain more information about the course of the extraction, I will use fluorescent labels to monitor the cellular content of critical components. First, I will monitor the loss of microfilaments using fluorescently-labeled Cytochalasin D (Cytochalasin D-BODIPY FL). Preliminary data has shown the cytochalasin-based stain successfully binding to Giardia. At low concentrations, Cyto D-BODIPY has been shown to leave cell morphology intact while labeling microfilaments (Munter et al., 2006). I will also attempt to monitor the the loss of cell membranes with membrane fluorescent membrane stains like CM-DiI and BODIPY FL C5-ceramide. This will allow me to judge when my treatments are successful in manipulating the presence of actin and the rest of the cell. Once I have sufficiently refined my treatment protocol, I will scale up the procedure and produce and purify quantities of actin. I will have to adapt the protocol to test tube-scale quantities in order to produce useful amounts of actin. This scaled-up extraction protocol will produce eluate enriched with actin, but still containing cell debris. Thus I will further treat this eluate to purify the actin for use. This may involve affinity chromatography using DNase I, as in the work of Patrón et al. (2005). Once actin has successfully been obtained, I will purify it, possibly through affinity The uses for a source of purified high-fidelity Giardia actin are numerous and diverse. The expression of a Toxoplasma gondii actin through baculovirus allowed characterization of its unusual kinetics and filament morphology (Sahoo et al., 2006). Also, a recent preparation of Plasmodium falciparum actin has yielded surprising data about its polymerization dynamics (Schüler et al., 2005). As an even more divergent eukaryote, it is impossible to predict how Giardia actin could behave. Electron microscopy studies could examine the inherent morphology of Giardial microfilament networks as they form spontaneously in vitro from pure actin (Sahoo et al., 2006). Knowing the natural structure and dynamics of Giardia's microfilament networks will help us elucidate how the cytoskeleton controls attachment. Also, no actin-binding proteins are currently known in Giardia (Elmendorf et al., 2003). This actin could be utilized to perform co-precipitation experiments to isolate proteins bound to microfilaments. This would allow us to better understand the actin dynamics and evolutionary history of this basal eukaryote. Drug binding studies could also be performed using a number of different methods, including surface plasmon resonance spectroscopy. Using this method, Giardial or human actin can be bound to a microscope slide and exposed to a pharmacophore, and affinity can be directly quantified (Dierynck et al., 2007). As emphasized above, this highly divergent protein presents a great opportunity to find safe drugs acting on this new target. Materials and Methods
Culturing Giardia lamblia
Genome strain Giardia identical to the isolate used for the Giardia Genome Project at the Marine Biological Laboratory (Woods Hole, MA) was used for these experiments. Cultures were maintained at 37°C in 9ml borosilicate tubes with TYI-S-33 medium (Keister, 1983) modified by replacing the phosphate buffer with 24mM sodium bicarbonate. Subculturing was carried out by putting a confluent 9ml tube (about 2x106 cells/ml) in an ice bucket for 15 minutes to detach the Giardia cells from the sides of the tube by cold. A varying amount of cold Giardia culture was inoculated into enough TYI- S-33 medium to fill the 9ml tube, leaving 0.5ml air. The amount of cold Giardia culture added varied from 0.25µl to 4ml. The volume was selected according to which would leave the tube 50% confluent, in mid-log growth phase, by the time of experiment, assuming a 6 hour doubling time. Tubes to perpetuate the culture more often were inoculated with 1-200µl. Preparation of Giardia for fluorescent staining and microscopic examination
In preparation for treatment with fluorescent stain and examination under light microscopy, a 10-well heavy Teflon-coated microscope slide was treated with 10µl 0.1% poly-L-lysine on each 5mm diameter well. After 15 minutes, the poly-L-lysine was washed off twice and replaced with Dulbecco's phosphate buffered solution (PBS). Then a 50% confluent Giardia culture tube was rinsed with 37°C PBS by replacing its volume twice. The tube was then incubated on ice for 15 minutes to detach Fluorescent labeling of Giardia microfilaments with Cytochalasin D-BODIPY FL
After incubating on ice for 15 minutes, about 1x106 cells/ml of PBS-suspended Giardia were treated with a final concentration between 10 and 50nM Cytochalasin D- BODIPY FL (Invitrogen, Inc.), except for a control aliquot. 50µl of the suspension was then placed in each well of the slide. The slide was then incubated at 37°C in 5% CO2 in a moistened container for between 45 and 60 minutes. In other trials, the Cyto D- BODIPY-treated suspension was incubated on ice for 25 minutes, then for 55 minutes at Fluorescent labeling of Giardia membranes with CM-DiI
In other trials, Giardia cells were labeled instead with CM-DiI (Invitrogen, Inc.). After the 15 minute incubation on ice, about 1x106 cells/ml of PBS-suspended Giardia were treated with a final concentration between 0.1 and 10µM DiI, except for a control aliquot. 50µl of the suspension was then placed in each well of the slide. In some trials, the DiI was added directly to the 50µl aliquots in each well. Then in some of those trials the slide and its aliquots were pre-warmed to 37°C. The slide was then incubated at 37°C in 5% CO2 in a moistened container for between 2.5 and 15 minutes. Fluorescent labeling of Giardia membranes with BODIPY FL C5-ceramide
In other trials, Giardia cells were labeled instead with BODIPY FL C5-ceramide (Invitrogen, Inc.). After the 15 minute incubation on ice, about 1x106 cells/ml of PBS- suspended Giardia were treated with a final concentration between of 200nM BODIPY- ceramide, except for a control aliquot. 50µl of the suspension was then placed in each well of the slide. Then the slide was incubated at 37°C in 5% CO2 in a moistened container for 30 minutes. Light microscope observation of detergent treatment of Giardia lamblia
Stained Giardia cells were treated with PHEM-NA buffer before observation under DIC and fluorescent microscopy. After the incubation with Cyto D-BODIPY, the fluid on the slide was replaced twice with PHEM-NA (60mM PIPES, 25mM HEPES, 10mM EGTA, 2mM MgCl2, 1mM ATP and a range of NP-40 between 2% and 40%) or PBS as a control. Then the slide was examined under a Zeiss Axioplan 2 microscope for periods between 20 to 90 minutes. Differential interference contrast imaging would be used, as well as fluorescence imaging. For Cyto D-BODIPY, and BODIPY-ceramide, the excitation wavelength used was 480nm and the emission wavelength used was 535nm. For DiI, the excitation wavelength used was 545nm and the emission wavelength used was 610nm. Microscopic observations of detergent treatment of Giardia with PHEM-NA
In order to begin refining the correct buffers and steps in detergent extracting actin from Giardia, I first devised a preliminary buffer: PHEM-NA. This is composed of PHEM, a common cytoskeleton-preserving buffer, ATP, required for continued actin polymerization, and NP-40, used in the cytoskeletal preparations of Benchimol et al. PHEM is composed of 60mM PIPES, 25mM HEPES, 10mM EGTA, and 2mM MgCl2. I then treated Giardia cells with PHEM-NA on a microscope slide in order to observe the progress of the treatment through differential interference microscopy (DIC). After treating a slide with poly-L-lysine to aid cellular attachment, I would allow cells to attach and then wash them twice with this buffer. When I first observed the results under the microscope, however, I only observed immobile cells detached from the slide for both the PBS-treated control and the PHEM-NA-treated cells. I was concerned that the amount of poly-L-lysine I used was excessive, therefore I then tested different applications of poly- L-lysine. One experiment showed that Giardia cells become immobile and detached from the slide when it is treated by allowing 50µl of 0.1% poly-L-lysine solution to dry on each 5mm diameter well before application of a suspension of Giardia cells. However, if each well on the slide is treated with 10µl of poly-L-lysine for 15 minutes before washing it off twice with PBS, Giarda cells will adhere to the slide when incubated at 37°C and their flagella will remain oscillating (data not shown). Presumably the lower volume of poly-L-lysine and the step of washing it off greatly reduced the concentration of poly-L-lysine in the solution containing Giardia cells. Resolving the correct application of poly-L-lysine allowed me to perform detergent treatments on cells not being adversely affected by another variable. I then proceeded to apply treatments of PHEM-NA with 2% - 40% NP-40, which subtly affected the appearance of the cells under DIC light microscopy. At first I used a range of detergent concentration in order to learn which is most effective at extracting the cells. Very quickly (within about 3 minutes) after treatment with any of the above NP-40 concentrations, Giardia cells appear flat under DIC microscopy, with much less contrast defining their features than in the PBS control (Fig. 1). After treatment with PHEM-NA with 10% NP-40, a small fraction of cells appear to lose the visible membrane encompassing the caudal complex. At this stage the outline of the cell seems to consist of the edges of the ventral disk and the caudal axonemes (Fig. 2). Not many cells appear in this state, though more appear as the detergent treatment progresses. By about 30 minutes of treatment, there are at most 1 cell out of every 12 in this state. Using DIC light microscopy, the observed shape and features of the cells are mostly retained during detergent treatment, with the exception of some cells whose shape is subtly altered. Fluorescent labeling of Giardia microfilaments with Cytochalasin D-BODIPY FL
Based on the DIC observations of only a change in contrast and shape, I was unable to make any definitive conclusions about the effects of PHEM-NA on the cell membranes or microfilaments. Thus I decided to stain those components of the cell before detergent treating in order to precisely visualize the effects on them. I first used Cytochalasin D-BODIPY FL, which consists of the BODIPY fluorophore bonded to Cytochalasin D, a fungal drug which normally binds to microfilaments and induces depolymerization. However, at low concentrations it can simply bind microfilaments, without disrupting them (Munter et al,. 2006). Thus at these concentrations it binds to microfilaments and fluorescently labels them. Initial trials using Cyto D-BODIPY successfully produced moderate fluorescence, but I could not obtain successful results beyond that point. When incubating in 50nM Cyto D-BODIPY for 55 min, the cells are capable of exhibiting a faint fluorescence that seems distributed throughout the cell (Fig. 3A). After some initially successful treatments, however, the results showed no fluorescence at all (Fig. 3B). After many experiments attempting to isolate the cause of the unsuccessful staining, I decided to put the troubleshooting on hold indefinitely. Fluorescent labeling of Giardia outer membranes with CM-DiI
In order to clarify the effects I observed earlier of PHEM-NA on the cell membranes, I began adapting for Giardia a protocol for fluorescently labeling the outer membranes of the cells with CM-DiI. DiI consists of a fluorophore and two hydrophobic hydrocarbon tails. The hydrophobic tails cause the molecule to incorporate into the plasma membrane of cells, thus labeling it with the attached fluorophore. In order to determine the most effective working concentration, I first performed an assay with serial dilutions of DiI staining cells at 37°C for 20 minutes. The results indicated that staining with 0.1µM DiI produces negligible staining of Giardia cells. 1µM DiI stains cells very rarely, and 10µM DiI stains at the most 1 in 6 cells. As a reference, the manufacturer protocol suggests a maximum working concentration of 5µM (Molecular Probes, Inc. 2005). When stained, the efficiency of the staining is extremely variable. Cells exposed to 10µM DiI sometimes fluoresce barely enough to be visible, while some cells exposed to 1µM DiI fluoresce brightly. Along with the inefficiency of DiI staining, when applied in the above manner it forms what seem to be aggregates in the medium. When applied in concentrations anywhere between 0.1µ M and 10µ M, DiI produces specks of very bright fluorescence separate from the cells. The effect can be observed after incubating 3.3 – 5µM DiI for anywhere between 2.5 and 20 minutes at 37°C. Pre-warming of the slide and cells does not block the effect. The size of the specks can range from smaller than a nucleus to a few cell-widths across. Some specks seem to associate with the outer membranes of stained cells, but most seem to be free-floating (Fig. 4). Because of the persistence of the inefficient staining, aggregation, and possible adverse effects on membrane integrity, I decided to investigate an alternative membrane dye. Fluorescent labeling of Giardia membranes with BODIPY FL C5-ceramide
In view of the difficulties inherent in membrane-staining with DiI, I moved to BODIPY-ceramide, a membrane dye based off of sphingosine. Sphingosine is a common membrane lipid, causing this molecule to also incorporate itself into membranes. This stain has been shown to label more evenly all the intracellular membranes in addition to the plasma membrane (Hernandez et al., 2007). I incubated the cells in 200nM BODIPY-ceramide for 30 minutes at 37°C, as described in Hernandez et al. (2007). This treatment produces a very efficient labeling, staining almost every cell (Fig. 5). Unfortunately, preliminary data suggested difficulties of BODIPY-ceramide to label cells after washing with PHEM-NA. After following the same procedure as above, I then washed the cells with PHEM-NA and control cells with PBS. No fluorescence was then observed in either PBS or PHEM-NA treatments (data not shown). However, I have performed too few detergent treatment trials with this stain to judge its usefulness in my study. The efficiency of its staining suggests optimism for its future use. Discussion
This project made some initial progress in refining a protocol to detergent extract Giardia cells to elute a sample enriched with actin. To begin, I observed some exploratory detergent treatments of cells on a microscope slide to gauge its effectiveness before scaling up. These observations showed progress, indicating that the extraction was achieving similar results to those in the papers by Benchimol et al. which it was based on. Using only DIC light microscopy it was impossible to verify that the observed events were the same as those depicted in the EM images by Benchimol et al. Thus I began working to refine the use of fluorescent labels to identify the presence of key parts of the cells. After encountering obstacles in the use of some labels, this work remains unfinished. But the initial results from the DIC observations suggest that my treatment may be appropriate only for the first phase of extraction, after which I may need to alter the procedure in order to elute actin in the proper way. Once eluted, the actin can then be purified in some method, for example with DNase I affinity chromatography (Patrón et al., 2005). This actin can then be used in a number of studies which require Giardia actin in its native conformation. Initially there was a problem with cells dying on the slide before the treatment, which turned out to be due to excessive concentrations of poly-L-lysine on the slide. My first procedure called for letting 50µl of 0.1% poly-L-lysine dry in each well, then adding PBS-suspended cells to the wells. This would re-solvate most of the poly-L-lysine again, resulting in a concentration which was apparently lethal to the cells. I discovered this in an experiment using decreasing amounts of poly-L-lysine. Smaller amounts of poly-L- lysine treatment resulted in more attached, animated cells. But more crucial was the step of washing off the poly-L-lysine with PBS before adding cell solution. Presumably this reduced the poly-L-lysine concentration greatly, indicating that the lethal effect was probably from a hypertonic solution. Through this experimentation I resolved this issue and thus could observe detergent treatments on cells unhindered by other effects. These experiments in detergent extracting Giardia cells with PHEM-NA initially showed little effect until I noticed a specific phenomenon which suggested partial lysis by the detergent. At first there seemed to be little extraction occurring, even with 40% detergent, since the cells seemed to mostly retain their intact morphology. But once I took into account the drastic change in contrast between treated and untreated cells, it became apparent that much was likely occurring. The phenomenon of cell ghosts has long been documented (Matter, 1979). A ghost is essentially the "shell" of a lysed cell which has lost its cytoplasm but retained most of its membrane and shape. This occurring in Giardia would be unsurprising, as its extensive cytoskeleton would cause its external morphology shape to persist. The cytoplasm normally creates contrast under DIC microscopy, as it has a different index of refraction than the surrounding medium. Ghosts, which lack cytoplasm and hence this boundary between different indices of refraction, will display borders not as well-defined. This is precisely what occurred in the detergent-treated cells. They lost the high contrast of their outer border exhibited by the PBS-treated cells. Their overall shape was usually preserved, with the teardrop outline of the cell intact (Fig. 1). This makes it unclear how extensive the membrane damage to the cells was. Loss of a relatively small patch of membrane could conceivably result in the extrusion of the cytoplasm. The other extreme possibility is that all of the dorsal membrane was removed, leaving only the part of the ventral surface attached to the slide. The tendency for this to occur is suggested by Benchimol (2005), where detailed SEM images show cells having lost all membranes except the ventral surface. These results were also obtained after a detergent extraction on live cells attached to a poly-L-lysine surface. Hence the loss of contrast in my cells made it appear that they were exhibiting the extensive loss of membrane and cell body shown in the paper by the Benchimol Another result in this article, fully-demembranated cells, seems to occur in my data, though the evidence is insufficiently conclusive. Though most cells retain their pre- treatment outline, after several minutes of detergent extraction I began to see cells which seemed to have lost all membranes (Fig. 2). The teardrop shape had been lost, seeming to leave only the ventral disk and caudal axonemes. In Benchimol (2005), there is an SEM image of a cell which has lost the ventral membrane and retains only the ventral disk, caudal axonemes, and nuclei after detergent treatment. This would create the same silhouette as I observed under DIC microscopy. As above, though, there are other possibilities as to what is actually occurring. From the observed data, it is also possible that the membrane immediately posterior to the ventral disk is simply being pulled in, flush with the disk and axonemes. Also, even if the appearance is due to membrane loss, it cannot be said whether the rest of the ventral membrane, which is hidden by the disk and axonemes, has also dissociated. The visual similarity of these results to those obtained in Benchimol (2005) makes it seem likely that they are the same phenomena. However, the paucity of information supplied by DIC light microscopy makes it necessary to confirm the actual course of events with more revealing methods. The use of membrane and microfilament fluorescent labeling was intended to provide specific information about the cellular content of components. Cells were to be stained with Cytochalasin D-BODIPY FL prior to detergent treatment, which would bind to their microfilaments to label them during the extraction. This would reveal how quickly and at what point microfilaments dissociate from the cell. Similarly, staining with CM-DiI or BODIPY-ceramide FL C5 would label membrane content during treatment and reveal the same information. Membrane labeling would reveal whether the cell membrane was actually present but pulled inward when the cells appeared to only retain the ventral disk and axonemes. BODIPY-ceramide, which stains more evenly the intracellular membranes as well as the plasma membrane, could provide additional information about the loss of a variety of other organelles. Thus these stains would hopefully resolve the question of whether my results were duplicating those in Benchimol Unfortunately, both Cyto D-BODIPY and DiI failed to stain cells efficiently, even through many alterations in protocol. Though Cyto D-BODIPY initially showed some successful results, it persistently failed to stain after those trials. The specks observed when staining with DiI suggest the amphipathic dye has a tendency to precipitate out of solution and aggregate with itself. After attempting a number of procedure variations, this tendency seems to be not easily overcome. This aggregation of the dye may or may not be responsible for the uneven and inefficient staining of the cells. The seemingly random staining of some cells and not others suggests some sort of stochastic process. This could be because the stain is sequestered in droplets, lowering the stain concentration in the medium to ineffective levels. It may be that a droplet will stain a cell when the proper contact is made, but this occurs infrequently, resulting in a few cells being relatively well stained (Fig. 4). However the mechanism, neither DiI nor Cyto D- BODIPY would efficiently stain using reasonable iterations on my procedure. In light of these difficulties, I moved to BODIPY-ceramide, which efficiently stained the periphery and cytoplasm of almost all treated cells. Unfortunately, though, initial experiments using BODIPY-ceramide during detergent treatment were unsuccessful. This may be because the PBS and PHEM-NA buffers replaced the medium containing the stain, if it is necessary for the cells to remain exposed to stain in order to fluoresce. Thus including BODIPY-ceramide in the PHEM-NA and PBS buffers may alleviate this problem. The use of BODIPY-ceramide may also be more useful than DiI beyond efficiency. BODIPY-ceramide stains more intracellular membranes than DiI, thus labeling intracellular organelles which represent a portion of the cytoplasmic content. This helps to reveal more information about when the extraction has eliminated cell debris, which hopefully will be before most of the microfilaments are disrupted. Hence BODIPY-ceramide shows much promise as a stain, though an unknown amount of work lies ahead before it can be made use of in a deteregent extraction. My most informative data remains the DIC imagery of ghosted cells, which unfortunately suggests that the timespan of actin dissociation may not be short enough to be useful. As described above, two crucial states observed during extractions seemed to represent ghosted cells and bare cytoskeletons with membranes wholly removed. These states likely represent two images in Benchimol (2005) where actin is present and then absent, respectively. If, during the course of detergent extraction, most cells became ghosts and then moved to the bare cytoskeleton stage in synchrony, a very useful actin fraction could be eluted. During the relatively short transition from the former stage to the latter, a high proportion of the material being eluted would be dissociating actin. This eluate would contain much more actin proportionally than a whole cell lysate would. Then this fraction could be further purified, perhaps using chromatography, and utilized in a number of studies requiring high-fidelity Giardia actin. But the span of time over which the cells seem to make this transition is almost the entire course of the extraction. Barring other solutions, this lack of rapid actin dissociation would drastically reduce the usefulness this detergent extraction. There are a number of actin-enriching alterations to the protocol which I could attempt. First, it is possible that better synchronization of actin dissociation could be obtained through a physical step such as agitation of the cells. If agitation fails, then I could treat with a different buffer which would cause a synchronized transition. After ghosting the cells with PHEM-NA, which includes ATP and Mg2+ to preserve actin, I could then treat with another buffer which would induce rapid dissociation of actin from the cytoskeleton. There was always a question whether obtaining the transition suggested in Benchimol (2005) would require simply longer treatment time, agitating the cells, or entirely switching detergent buffers. If these alterations are unsuccessful, I could take an alternative strategy which does not attempt to dissociate actin from the cytoskeleton at all. This would involve using dimethyl sulfide to crosslink microfilaments to the cytoskeleton and saving this as a cytoskeleton-enriched fraction. This would be similar to the approach of Patrón et al. (2005) in obtaining a T. gondii subpellicular network fraction enriched with actin before further purifying the actin through chromatography. The future progress to be made in this project involves the above steps in refining the extraction, then scaling up the extraction and purifying the actin product. First I have to resolve, on the microscope slide, whether the cells require a physical or chemical prompt to rapidly relinquish their actin. Then I must make the proper adjustments to translate this knowledge into a production-scale procedure. Even a treatment using a few milliliters of cell suspension could have very different behavior. In the course of producing this procedure I would use some diagnostic to visualize or quantify the amount and purity of actin I elute. A simple SDS-PAGE gel could give an indication of this. Then I must further purify the actin obtained, likely using a similar affinity chromatography setup to that in Patrón et al. (2005). They utilized the affinity of actin for the enzyme DNase I to immobilize actin on a DNase I column. It is largely unknown whether the unique actin of Giardia will efficiently bind DNase I, though very preliminary data from our lab suggests that it can. There are many steps before I have a full procedure to obtain pure, native Giardia actin, but that actin would enable many extremely informative studies. High-fidelity Giardia actin would greatly enable studies on the cytoskeletal system of Giardia and experiments on drug targets in the parasite. The basic dynamics of giardial microfilaments, their networks, and regulation are currently completely uncharacterized. The key obstacle to studying these issues is obtaining quality Giardia actin. After similarly purifying T. gondii actin, Patrón et al. (2005) were able to make morphological characterizations about the microfilaments of that species using electron microscopy. Another potential achievement would be the identification of an actin- binding protein in Giardia. Using Giardia G-actin and F-actin as the "bait" in affinity chromatography, it may be possible to extract actin binding proteins from Giardia lysate. The existence and exact nature of these proteins would reveal much about the unique system of regulating microfilament dynamics in Giardia. These proteins would presumably be unique to Giardia and also would mediate actin dynamics, a process crucial in attachment. This suggests them as potential drug targets, alongside actin itself. Using actin and any associated proteins, we may perform drug binding studies using surface Plasmon resonance microscopy. This allows the quantification of drug affinity to Giardia and human actin, along with other important drug kinetics. Importantly, a study by Dierynck et al. (2007) shows that only a few changes in a protein are required to change the values of these kinetics by several orders of magnitude. Thus studying actin expressed exogenously in E coli., which is known to lack required folding mediators (Vainberg et al., 1998; Speiss et al., 2004), risks invalidating the quantification. The same concerns apply to all of the aforementioned potential quantifications and characterizations. Thus the effort to obtain high quality Giardia actin, through this or any other means, is crucial to understanding the cytoskeleton of Giardia lamblia and the disease it causes. References
Ashbolt, N.J. 2004. Microbial contamination of drinking water and disease outcomes in developing regions. Toxicology. 198:229-238 Astiazarán-García, H., Espinosa-Cantellano, M., Castanón, G., Chávez-Munguía, B., and A. Martínez-Palomo. 2000. Giardia lamblia: effect of infection with symptomatic and asymptomatic isolates on the growth of gerbils (Meriones unguiculatus). Experimental Parasitology. 95:128-135 Benchimol, M. 2005. The nuclei of Giardia lamblia —new ultrastructural observations. Archives of Microbiology 183:160-168 Benchimol, M., Piva, B., Campanati, L., and W. de Souza. 2004. Visualization of the funis of Giardia lamblia by high-resolution field emission scanning electron microscopy—new insights. Journal of Structural Biology. 147:102-115 Boucher S.M. and F.D. Gillin. 1990. Excystation of in vitro-derived Giardia lamblia cysts. Infection and Immunity. 58:3516-3522 Bryce, J., Boschi-Pinto, C., Shibuya, K., Black, R.E., and WHO Child Health Epidemiology Reference Group. 2005. WHO estimates of the causes of death in children 365:1147-1152 Carvalho, K.P. and L.H. Monteiro-Leal. 2004. The caudal complex of Giardia lamblia and its relation to motility. Experimental Parasitology. 108:154-162 céu Sousa, M., Goncalves, C.A., Bairos, V.A., and J. Poiares-da-Silva. 2001. Adherence of Giardia lamblia Trophozoites to Int-407 Human Intestinal Cells. Clinical and Diagnostic Laboratory Immunology. 8:258-265 Charras, G.T., Hu, C.K., Coughlin, M., and T.J. Mitchison. 2006. Reassembly of contractile actin cortex in cell blebs. Journal of Cell Biology. 175:477-90. Correa, G., Morgado-Diaz, J.A., and M. Benchimol. 2004. Centrin in Giardia lamblia - ultrastructural localization. FEMS Microbiology Letters. 233:92-96 Correa, G., and M. Benchimol. 2006. Giardia lamblia behavior under cytochalasins treatment. Parasitology Research. 98:250-256 Crossley, R., and D. Holberton. 1985. Assembly of 2-5 nm filaments from giardin, a protein associated with cytoskeletal microtubules in Giardia. The Journal of Cell Science. 78:205-231 Dacks, J.B., Davis, L.A.M., Sjogren, A.M., Andersson, J.O., Roger, A.J., and W.F. Doolittle. 2003. Evidence for Golgi bodies in proposed ‘Golgi-lacking' lineages. Proceedings of the Royal Society of London Series B (Supplement). 270:S168- de Souza, W., Campanati, L., and M. Attias. 2008. Strategies and results of field emission scanning electron microscopy (FE-SEM) in the study of parasitic protozoa. Micron. 39:77-87 Dierynck, I., De Wit, M., Gustin, E., Keuleers, I., Vandersmissen, J., Hallenberger, S., and K. Hertogs. 2007. Binding kinetics of darunavir to human immunodeficiency virus type 1 protease explain the potent antiviral activity and high genetic barrier. Journal of Virology. 81:13845-13851 Doolittle, R.F., and A.L. York. 2002. Bacterial actins? An evolutionary perspective. BioEssays. 24:293-296 Drouin, G., Moniz de Sá, M., and M. Zuker. 1995. The Giardia lamblia Actin Gene and the Phylogeny of Eukaryotes. Journal of Molecular Evolution. 41:841-849 Elmendorf, H.G., Dawson, S.C., and J.M. McCaffery. 2003. The cytoskeleton of Giardia lamblia. International Journal for Parasitology. 33:3-28 Emelyanov, V.V. 2003. Phylogenetic affnity of a Giardia lamblia cysteine desulfurase conforms to canonical pattern of mitochondrial ancestry. FEMS Microbiology Letters. 226:257-266 Erlandsen, S.L., Russo, A.P., and J.N. Turner. 2004. Evidence for Adhesive Activity of the Ventrolateral Flange in Giardia lamblia. The Journal of Eukaryotic Microbiology. 51:73-80 Erlandsen, S.L., and D.G. Chase. 1974. Morphological alterations in the microvillous border of villous epithelial cells produced by intestinal microorganisms. The American Journal of Clinical Nutrition. 27:1277-1286 Farthing, M.J.G. 1997. The molecular pathogenesis of giardiasis. Journal of Pediatric Gastroenterology and Nutrition. 24:79-88 Feely, D.E., and S.L. Erlandsen. 1982. Effect of cytochalasin-B, low Ca++ concentration, iodoacetic acid, and quinacrine-HCl on the attachment of Giardia trophozoites in vitro. The Journal of Parasitology. 68:869-73 Feely, D.E., Schollmeyer, J.V., and S.L. Erlandsen. 1982. Giardia spp.: distribution of contractile proteins in the attachment organelle. Experimental Parasitology. Flaiz, C., Kaempchen, K., Matthies, C., and C.O. Hanemann. 2007. Actin-rich protrusions and nonlocalized GTPase activation in Merlin-deficient schwannomas. Journal of Neuropathology and Experimental Neurology. 66:608- Friend, D.S. 1966. The fine structure of Giardia muris. The Journal of Cell Biology. Fox, L.M., and L.D. Saravolatz. 2005. Nitazoxanide: A New Thiazolide Antiparasitic Agent. Reviews of Anti-Infective Agents. 40:1173-1180 Gillin, F.D., and D.S. Reiner. 1982. Attachment of the flagellate Giardia lamblia: role of reducing agents, serum, temperature, and ionic composition. Molecular and Cell Biology. 2:369-77 Gillin, F.D., Reiner, D.S., and J.M. McCaffery. 1996. Cell biology of the primitive eukaryote Giardia lamblia. Annual Reviews of Microbiology. 50:679-705 Hanevik, K., Hausken, T., Morken, M.H., Strand, E.A., Mørch, K., Coll, P., Helgeland, L., and N. Langeland. 2007. Persisting symptoms and duodenal inflammation related to Giardia duodenalis infection. The Journal of Infection. 55:524-530 Hansen, W.R., Tulyathan, O., Dawson, S.C., Cande, W.Z., and D.A. Fletcher. 2006. Giardia lamblia attachment force is insensitive to surface treatments. Eukaryotic Holberton, D.V. 1973. Fine structure of the ventral disk apparatus and the mechanism of attachment in the flagellate Giardia muris. Journal of Cell Science. 13:11-41 Holberton, D.V. 1974. Attachment of Giardia – a hydrodynamic model based on flagellar activity. The Journal of Experimental Biology. 60:207-221 Holberton, D.V. 1981. Arrangement of subunits in microribbons from Giardia. The Journal of Cell Science. 47:167-185 Holberton, D., Baker, D.A., and J. Marshall. 1988. Segmented alpha-helical coiled-coil structure of the protein giardin from the Giardia cytoskeleton. Journal of Molecular Biology. 204:789-795 Katelaris, P.H., Naeem, A., and M.J. Farthing. 1995. Attachment of Giardia lamblia trophozoites to a cultured human intestinal cell line. Gut. 37:512-8 Keister, D.B. 1983. Axenic culture of Giardia lamblia in TYI-S-33 medium supplemented with bile. Transcriptions of the Royal Society of Tropical Medicine and Hygiene. 77:487-488 Khoury, R.S. 2004. The Role of Microfilaments in the Attachment of Giardia lamblia: a Cytoskeletal Drug Study. Georgetown University (thesis) Lane, S., and D. Lloyd. 2002. Current trends in research into the waterborne parasite Giardia. Critical Reviews in Microbiology. 28:123-147 Luján, H.D., Mowatt, M.R., Byrd, L.G., and T.E. Nash. 1996. Cholesterol starvation induces differentiation of the intestinal parasite Giardia lamblia. Proceedings of the National Academies of Sciences. 93:7628-7633 Mandomando, I.M., Macete, E.V., Ruiz, J., Sanz, S., Abacassamo, F., Valles, X., Sacarlal, J., Navia, M.M., Vila, J., Alonso, P.L., and J. Gascon. 2007. Etiology of diarrhea in children younger than 5 years of age admitted in a rural hospital of southern Mozambique. The American Journal of Tropical Medicine and Hygiene. Marshall, J., and D.V. Holberton. 1993. Sequence and structure of a new coiled coil protein from a microtubule bundle in Giardia. Journal of Molecular Biology. Matter, A. 1979. Microcinematographic and electron microscopic analysis of target cell lysis induced by cytotoxic T lymphocytes. Immunology 36:179-190 McCabe, R.E., Yu, G.S., Conteas, C., Morrill, P.R., and B. McMorrow. 1991. In vitro model of attachment of Giardia intestinalis trophozoites to IEC-6 cells, an intestinal cell line. Antimicrobial Agents and Chemotherapy. 35:29-35 Munter, S., Enninga, J., Vazquez-Martinez, R., Delbarre, E., David-Watine, B., Nehrbass, U., and S.L. Shorte. 2006. Actin polymerisation at the cytoplasmic face of eukaryotic nuclei. BMC Cell Biology. 7:23 Narcisi, E.M., Paulin, J.J., and M. Fechheimer. 1994. Presence and localization of vinculin in Giardia. Journal of Parasitology. 80:468-473 Nozaki, T., and V. Ali. 2007. Current Therapeutics, Their Problems, and Sulfur-Containing- Amino-Acid Metabolism as a Novel Target against Infections by "Amitochondriate" Protozoan Parasites. Clinical Microbiology Reviews. 20:164-187 Oliveira-Lima, J.G., Mineo, J.R., Santos, A.A., Ferreira, G.L., Ferro, E.A., and C.A. Oliveira. 2001. Improved methods for examination of Toxoplasma gondii cytoskeleton at ultrastructural level. Parasitology Research. 87:287-293 Olson, M.E., O'Handley, R.M., Ralston, B.J., McAllister, T.A., and R.C.A. Thompson. 2004. Update on Cryptosporidium and Giardia infections in cattle. Trends in Parasitology. 20:185-191 Patrón, S.A., Mondragón, M., González, S., Ambrosio, J.R., Guerrero, B.A.L., and R. Mondragón. 2005. Identification and purification of actin from the subpellicular network of Toxoplasma gondii tachyzoites. International Journal for Parasitology. Peattie, D.A., Alonso, R.A., Hein, A., and J.P. Caulfield. 1989. Ultrastructural Localization of Giardins to the Edges of Disk Microribbons of Giarida lamblia and the Nucleotide and Deduced Protein Sequence of Alpha Giardin. The Journal of Cell Biology. 109:2323-2335 Petri, W.A. 2003. Therapy of intestinal protozoa. Trends in Parasitology. 19:523-526 Piva, B., and M. Benchimol. 2004. The median body of Giardia lamblia: an ultrastructural study. Biology of the Cell. 96:735-746 Ralston. B.J., McAllister, T.A., and M.E. Olson. 2003. Prevalence and infection pattern of naturally acquired giardiasis and cryptosporidiosis in range beef calves and their dams. Veterinary Parasitology. 114:113-122 Roskens, H., and S.L. Erlandsen. 2002. Inhibition of in vitro attachment of Giardia trophozoites by mucin. Journal of Parasitology. 88:869-873 Sant'Anna, C., Campanati, L., Gadelha, C., Lourenco, D., Labati-Terra, L., Bittencourt- Silvestre, J., Benchimol, M., Cunha-e-Silva, N.L., and W. de Souza. 2005. Improvement on the visualization of cytoskeletal structures of protozoan parasites using high-resolution field emission scanning electron microscopy (FESEM). Histochemistry and Cell Biology. 124:87-95 Schatten, H., Sibley, L.D., and H. Ris. 2003. Structural evidence for actin-like filaments in Toxoplasma gondii using high-resolution low-voltage Field Emission Scanning Electron Microscopy. Microscopy and Microanalysis. 9:330-335 Shapiro, M. 2006. A Bioinformatics Examination of the Cytoskeleton of Giardia lamblia. Georgetown University (thesis) Sogin, M.L., Gunderson, J.H., Elwood, H.J., Alonso, R.A., and D.A. Peattie. 1989. Phylogenetic Meaning of the Kingdom Concept: An Unusual Ribosomal RNA from Giardia lamblia. Science. 243:75-77 Speiss, C., Meyer, A.S., Reissmann, S., and J. Frydman. 2004. Mechanism of the eukaryotic chaperonin: protein folding in the chamber of secrets. Trends in Cell Biology. 14:598-604 Vainberg, I.E., Lewis, S.A., Rommelaere, H., Ampe, C., Vandekerckhove, J., Klein, H.L., Cowan, N.J. 1998. Prefoldin, a Chaperone that Delivers Unfolded Proteins to Cytosolic Chaperonin. Cell. 93:863-873 Welch, M.D., and R.D. Mullins. 2002. Cellular control of actin nucleation. Annual Review of Cell and Developmental Biology. 18:247-88 Yin, J., Li, G., Ren, X., and G. Herrler. 2007. Select what you need: A comparative evaluation of the advantages and limitations of frequently used expression systems for foreign genes. Journal of Biotechnology. 127:335-347 Figure 1. Detergent treatment with of Giardia lamblia attached to a poly-L-lysine
slide. Image captured using DIC light microscopy at 40X magnification. (A) cells treated with PHEM-NA containing 2% NP-40 for 17 minutes. (B) control cells treated with PBS for 22 minutes. Figure 2. Detergent treatment of Giardia lamblia attached to a poly-L-lysine slide.
Image captured using DIC light microscopy at 63X magnification. Cells were treated with PHEM-NA containing 10% NP-40 for 18 minutes. Inset: center cell enlarged a further 2X, with digitally enhanced contrast. Figure 3. Giardia lamblia attached to a poly-L-lysine treated slide, treated with 50nM
Cytochalasin D-BODIPY FL for 55 minutes at 37°C. Images captured using fluorescent microscopy at an excitation wavelength of 480nm and an emission wavelength of 535nm. (A) an early trial, captured at 20X magnification and digitally enhanced in brightness and contrast. (B) a later trial, captured at 63X magnification. Figure 4. Giardia lamblia cells attached to a poly-L-lysine treated slide, treated with
10µM CM-DiI for 15 minutes at 37°C. Image captured at 63X magnification using fluorescent microscopy at an excitation wavelength of 545nm and an emission wavelength of 610nm. Figure 5. Giardia lamblia cells attached to a poly-L-lysine treated slide, treated with
200nM BODIPY FL C5-ceramide for 30 minutes at 37°C. Images obtained at 63X magnification focused on the same field of view. Three cells are visible at the upper left, middle, and bottom right corner of the images. (A) fluorescent microscopy at an excitation wavelength of 480nm and an emission wavelength of 535nm. Contrast and brightness have been digitally enhanced. (B) DIC light microscopy.

Source: http://www.nsto.co/media/thesis.pdf