1 ANTHRAX LETHAL TOXIN PARALYZES ACTI N-BASED MOTILITY BY BLOCKING Hsp27 PHOSPHORYLATION By RUSSELL LAVON DURING, Jr. A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA
2 Copyright 2006 by Russell Lavon During, Jr.
3 To my pare nts and grandmother, who always supported my educational endeavors.
4 ACKNOWLEDGMENTS I would like to thank the me mbers of my committee for th eir support and guidance during my graduate studies. I would al so like to thank my mentor fo r the wonderful instruction I received during my PhD studies . His daily encouragement and enthusiasm for science were greatly appreciated. I must also thank fellow la b members for the techni cal guidance I received during my training. Without them this work would have not been possible. Finally, I must thank my parents for their love and support during my education.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES................................................................................................................ .........9 ABSTRACT....................................................................................................................... ............10 CHAPTER 1 INTRODUCTION................................................................................................................. .12 Bioterrisom and Bacillus anthracis ........................................................................................12 History and B. anthracis Microbiology...........................................................................12 Anthrax Infections...........................................................................................................13 Anthrax Toxins................................................................................................................14 MAPK Signaling Pathway......................................................................................................15 p38 MAPK....................................................................................................................... .......16 Macrophage Apoptosis...........................................................................................................17 Neutrophil Function............................................................................................................ ....18 Neutrophil Chemotaxis.......................................................................................................... .22 Actin Structure and Function..................................................................................................23 Actin Binding Proteins......................................................................................................... ..24 Listeria mononcytogenes ........................................................................................................25 LT Paralyzes Neutrophil Motility by Blocking Hsp27 Phosphorylation...............................27 2 MATERIALS AND METHODS...........................................................................................34 Neutrophil Motility and Actin Assembly...............................................................................34 Toxin Purification............................................................................................................34 Neutrophil Isolation, Toxin Trea tment and MAPKK Western Blot...............................34 Neutrophil Annexin V Staining, Propidi um Iodide Staining and NBT Test...................35 Neutrophil Chemokinesis, Chem otaxis, and Polarization...............................................35 Neutrophil Phalloidin Staining........................................................................................36 Triton X-100 Insoluble Cytoskeleton..............................................................................36 LT and Hsp27................................................................................................................... ......37 2D SDS-PAGE and Hsp 27 Western Blotting.................................................................37 Listeria Infection and Phalloidin Staining.......................................................................38 Recombinant Hsp27........................................................................................................39 Listeria Motility in Rat Brain Extracts............................................................................39 RNA Interference............................................................................................................40 Neutrophil and Listeria Immunoflorescence...................................................................40 Confocal Immunofluorescence of Neutrophils...............................................................41 Actin Kinetics................................................................................................................. ........41
6 Monomer Sequestering....................................................................................................41 Barbed End Assembly With F-Actin Seeds....................................................................42 Barbed End Actin Assembly With Spectrin 4.1 Seeds....................................................43 Critical Concentration Assays.........................................................................................43 Disassembly Assays........................................................................................................43 Monomer Sequestration After Treat ment of Hsp27 With MAPKAP-2..........................44 Statistical Analysis........................................................................................................... .......44 3 LITERATURE REVIEW.......................................................................................................45 Anthrax Clinical Data.......................................................................................................... ...45 Previous Studies with LT and Neutrophils.............................................................................46 4 RESULTS...................................................................................................................... .........47 Anthrax Toxins Paralyze Neutrophil Actin Assembly...........................................................47 Anthrax Lethal Toxin Does Not Cause Apoptosis or Necrosis in Human Neutrophils...................................................................................................................47 Anthrax Lethal Toxin Impairs Neutr ophil Chemotaxis and Chemokinesis....................48 Anthrax Lethal Toxin Blocks Neutrophil Actin Assembly.............................................50 p38 MAPK is Required For Neutrophil Actin Assembly in Response to FMLP...........51 LT Blocks Hsp27 Phosphorylation.........................................................................................52 Anthrax Lethal Toxin and the p38 Inhibitor SB203580 Bl ock Intracellular Listeria monocytogenes Actin Assembly..................................................................................52 Anthrax LT Blocks Phosphorylation of Hsp27 in HeLa Cells and Human Neutrophils...................................................................................................................53 Hsp27 Immunoflouresence..............................................................................................54 Non-phosphorylated Hsp27 Inhibits Listeria Actin Tail Formation In Rat Brain Extracts....................................................................................................................... .56 Hsp27 RNAi Blocks LT Activity....................................................................................56 Actin Kinetics................................................................................................................. ........57 Hsp27 is an Actin Monomer Sequestering Protein.........................................................57 MAPKAP-2 Reverses The Ability of Hsp27 to Sequester Actin Monomers.................58 Non-Phosphorylated Hsp27 Weakly Reduces Barbed End Actin Assembly..................58 Hsp27 AA and EE Slow Disassembly of Actin Filaments..............................................59 5 DISCUSSION................................................................................................................... ......76 LT Paralyzes Neutrophil Actin Assembly..............................................................................76 Anthrax Lethal Toxin Blocks Listeria monocytogenes Intracellular Motility........................79 Hsp27 Function................................................................................................................. ......81 Hsp27 and Neutrophils.......................................................................................................... .83 Kinetic Analysis of Hsp27 with Actin....................................................................................84 6 FUTURE WORK.................................................................................................................. ..90 The Effects of Edema Toxin on Neutrophils..........................................................................90 The Effect of LT Macrophages, De ndritic Cells and Platelets...............................................90
7 Role of Hsp27 and MAPK in Listeria and Neutrophil Motility.............................................91 Mutational Analysis of Hsp27................................................................................................92 LIST OF REFERENCES............................................................................................................. ..94 BIOGRAPHY...................................................................................................................... ........107
8 LIST OF TABLES Table page 4-1 Percentage of necrotic and apoptotic neutrophils..............................................................61 4-2 Densiometry scans of HeLa and Neutrophil 2D-western blots.........................................61
9 LIST OF FIGURES Figure page 1-1 Mitogen activated protein kina se (MAPK) signaling pathways........................................28 1-2 The p38 MAPK pathway................................................................................................. ..29 1-3 The neutrophil FMLP receptor signaling pathway............................................................30 1-4 The growth curve for actin assembly.................................................................................31 1-5 Actin filament treadmilling................................................................................................32 1-6 L. monocytogenes uses the host cells actin machinery......................................................33 4-1 Western blot showing MA PKK-1 N-terminal breakdown................................................61 4-2 Anthrax LT blocks neutrophil chemokinesis.....................................................................62 4-3 Anthrax LT blocks neutrophil chemotaxis........................................................................63 4-4 Anthrax LT blocks neutrophil actin assembly....................................................................64 4-5 p38 MAPK is required for neutrophil actin assembly.......................................................65 4-6 Anthrax LT blocks intracellular Listeria actin assembly...................................................66 4-7 2D-SDS PAGE of control He La cells and LT (1g/ml) treated cells...............................67 4-8 LT blocks Hsp27 phosphorylation ...................................................................................68 4-9 Hsp27 localizes to Listeria actin tails and the leading edge of neutrophils......................69 4-10 Confocal immunofluoresence of polarized neutrophils.....................................................70 4-11 Hsp27 siRNA blocks the action of LT...............................................................................71 4-12 Hsp27 sequesters actin monomers.....................................................................................72 4-13 Hsp27 weakly sequesters actin monomers when barbed ends are free.............................74 4-14 Hsp27 slows the disassembly of actin filaments................................................................75 5-1 Diagrammatic representation of Hsp27 function...............................................................89
10 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ANTHRAX LETHAL TOXIN PARALYZES AC TIN-BASED BY BLOCKING Hsp27 PHOSPHORYLATION By Russell Lavon During, Jr. December 2006 Chair: Frederick S. Southwick Major Department: Medical Sc iences--Immunology and Microbiology Bacillus anthracis causes high-level bacteremia, str ongly suggesting paralysis of the innate immune system. We have examined the e ffects of anthrax lethal toxin (LT) on human neutrophil chemotaxis, a process that requires acti n filament assembly. Neutrophils treated with a sub-lethal concentration of LT (50 ng/ml) for 2 h demonstrated insignificant apoptosis or necrosis. However, this same concentrati on slowed human neutrophil FMLP (formylated methionyl-leucyl-phenylanine)-stimulated chemok inesis by over 60%, markedly reduced polar morphology, and rendered neutrophil s incapable of responding to a chemotactic gradient. These changes were accompanied by a greater than 50% reduction in FMLP-induced actin filament assembly. One-hour exposure to LT failed to impair polarity or actin assemb ly, and the effects of LT were independent of MAPKK-1/2 (mitogen activ ated protein kinase kinases) inhibition. I have also found that LT paralyzes intracellular Listeria monocytogenes actin-based motility. Lethal toxin disrupts the p38 MAPK (mitogen act ivated protein kinase) pathway and blocks Hsp27 (heat shock protein 27) phosphorylation. Unphosphorylated Hsp27 concentrates in the actin tails of Listeria and the lamellipodia of neutrophils, inhibits Listeria tail formation in extracts, and blocks the assembly of purified actin. Unphosphorylated Hsp27 was found to be a potent monomer sequestering protein when the barbed ends of actin filaments are capped.
11 Phosphorylation of Hsp27 reverses all of thes e effects. The p38 inhibitor SB2035850 blocked neutrophil actin assembly and Listeria motility, further indicating the importance of this pathway for actin assembly. These findings provide a previously unappreciated mechanism for LT virulence and emphasize a central role for th e p38 MAPK pathway and Hsp27 in actin-based motility.
12 CHAPTER 1 INTRODUCTION Bioterrisom and Bacillus anthracis Bacillus anthracis infection continues to be a major concern in the United States because of its effectiveness as a bioterriosm agent. In 1970, the World Health Organization (WHO) estimated that if 50 kg of spores were releas ed, 200,000 people could be killed or rendered seriously ill . In 2001, anthrax spores were used successfully in the U.S as a bioterroist agent. Mailing anthrax spores through the mail successfully infected twenty-two people . There were 11 cases of cutaneous anthrax and 11 cases of inhalation anthrax reported. Death occurred in 5 out of 11 inhalation anthrax cases, with no deaths being reported for any of the cutaneous cases . The bioterriost attacks of 2001 have renewed interest in anthrax basic science research. By improving our knowledge of anthrax pathogenesis and other aspects of anthrax physiology, we hope to save lives by developing new means to fi ght anthrax infection, and by developing better diagnostic tests to allow physicians to rapidly detect anthrax exposure. History and B. anthracis Microbiology Anthrax is thought to have been the fifth plague that killed Egyptian cattle in the biblical Book of Exodus. Kochâ€™s work on anthrax is what eventually led to Kochâ€™s postulates. These postulates were used to prove that a particul ar organism caused a sp ecific disease state and required isolation of the organism and reinfection of another host. Anthrax is a zoonotic infection that primarily affects herbivores such as sheep and cattle . It is through contact with herbivores that humans usually acquire natural infections. Bacillus anthracis is a gram-positive rod organism that forms boxcar shaped chains wh en grown in culture . Anthrax can be grown on sheep blood agar, and the growth pattern is said to take on the shape of Medusaâ€™s head. Polymerase chain reaction and lysis by -phage are two methods that are used to diagnose the
13 presence of anthrax. Anthrax has two important virulence factors that are important for the infectious process. B. anthracis produces a poly-glutamic acid capsu le that is anti-phagocytic and also produces three exotoxins that are able to inhibit the host immune response [4,5]. These toxins are known as edema toxi n (ET) and lethal toxin (LT). B. anthracis has two virulence plasmids that encode for the poly-glutamic acid capsule and the toxins respectively. The plasmid pX01 encodes for both edema toxin and lethal toxin, while plasmid pX02 encodes for the polyglutamic acid capsule [6,7]. Deoxoyribonucleic acid (DNA) microarray anal ysis demonstrated extensive homology between chromosomal genes of B. anthracis and B. cereus . B. anthracis has 66-92% homology with different B. cereus strains when considering only chromosomal genes . DNA microarray experiments also found extensive ho mologs of genes residing on the virulence plasmid pX01 in B. cereus strains that were analyzed . The pathogenicity is land that encodes for anthrax toxins on pX01 was missing in all B. cereus strains that were st udied. This indicates that anthrax toxins are unique to B. anthracis and are not found in B. cereus strains. Unlike pX01, few homologous genes were found in B. cereus that corresponded with genes found on pX02 . This may explain why only B. anthracis produces a poly-glutam ic acid capsule and none of the other Bacillus species have this capability. Anthrax Infections Anthrax infection can present in three different forms depending on the site of entry into the host. Inhalation anthrax results when spor es are inhaled by infected individuals, and primarily results in a severe pneumonia-like syndrome that often results in high mortality (95%) . Once the spores are inhaled into the lungs , alveolar macrophages pha gocytose the spores and carry them to the mediastinal lymph nodes wher e germination and subsequent rupture of the macrophages ensues, resulting in the release of ve getative anthrax into the blood stream .
14 Significant pulmonary infiltrates, pleural effusi ons, and meningitis are often noted in patients with severe disease, although death is usually ca used by sepsis and respiratory failure . Cutaneous anthrax results when spores penetr ate the skin of infect ed individuals through sores or other breaks in the skin barrier. Lo cal germination occurs and results in eschar formation with rarely any systemic involvement. Cu taneous anthrax is rarely fatal (1%) if treated with antibiotics . Gastrointestinal anthrax usually occurs when anthrax spores are consumed by individuals who eat meat contaminated with th e spores. Abdominal pain, nausea, and vomiting are early symptoms, but if left untreated can result in bloody dia rrhea and eventually sepsis . Localized gastrointestin al lesions result, but systemic i nvolvement and mortality (40-50% if untreated) are also much less than that seen in inhalation anthrax if treated by antibiotics . Anthrax Toxins Anthrax pathogenesis is mediated by a tripartit e toxin that consists of protective antigen (PA), lethal factor (LF), a nd edema factor (EF). Protectiv e antigen is responsible for translocating LF and EF into the cytoplasm through a pH dependant endocytic event . Protective antigen has been shown to bind to cell surface receptors called tumor endothelial marker 8 (TEM-8) and the recently discovere d capillary morphogenesis protein 2 (CMG-2) [11,12]. Protective antigen, once bound, becomes cleaved by furin, which is present in the outer membrane. Protective antigen is cleaved from 83 -kD to 63-kD and 20kD fragments . The 63kD fragment is the active form, which then clusters to form heptamers so that LF or EF is able to bind . Heptamer formation is required for EF or LF binding. Up to three molecules of LF or EF can bind one heptamer of PA . Protective antigen + lethal factor is referred to lethal toxin (LT) and PA+EF is known as edem a toxin (ET). Lethal factor is a Zn2+ dependant metalloprotease that has previous ly been shown to have specific ity for mitogen activated protein
15 kinase kinases 1-4 and 6/7 (MAPKK) . Leth al toxin cleaves the N-terminal seven amino acids (phosphorylation domain) fr om MAPKK, which results in their inactivation . Edema factor functions as an adenyl ate cyclase that converts ATP (adenosine triphosphate) to cAMP (cyclic-adenosine monophosphate) . A rise in cAMP levels in neutrophils inhibits phagocytosis of the avirul ent Sterne strain of B. anthracis . MAPK Signaling Pathway There are three important mitogen activated pr otein kinase (MAPK) pathways (Fig. 1-1): p38, which becomes activated during times of stress or by inflammatory cytokines; ERK (extracellular signal-regula ted kinase), which is responsible for cell growth and proliferation; and finally JNK (c-Jun N-terminal kinase), which is also important for cell growth and stress responses . Anthrax LT can disable all th ree of the MAPK pathways . The MAPK signaling pathway is activated by membrane bound receptors, such as G-protein coupled receptors and tyrosine kinase receptors that beco me activated in response to substrate binding. In the case of G-protein coupled re ceptors, ligand binding causes the activation of the associated Gprotein, by facilitating the ex change of GDP (guanosine di phosphate) for GTP (guanosine triphosphate) in the subunit of the G-protein [2 0,21]. Once this occurs the subunit is thought to cause the activation of mitogen-activated pr otein kinase kinase ki nases (MAPKKK) that phosphorylate mitogen-activated protein kina se kinases (MAPKK), and that in turn phosphorylate mitogen-activated protein kinase s (MAPK) [20,21]. MAPK are phosphorylated on both tyrosine and threonine resi dues leading to their activation . Once activated, the MAPK, such as p38 can then phosphorylate effector protei ns on serines and threonines, leading to their activation . The exact mechanism of MAPKKK activation by the subunit of the Gprotein is not well understood, but small GTPa ses are thought to play a role [20,21].
16 The activation of MAPK th rough tyrosine kinase signaling is well understood [23,24,25]. One example is the activation of Erk by tyrosine kinase receptors. Once the substrate molecule binds, the tyrosine kinase receptor autophosphorylat es itself. In the case of Erk signaling, this would lead to the activation of Ras GEF (guanos ine exchange factor). Ras GEF catalyzes the exchange of GDP for GTP in the small GTPase Ras . Ras then activates the MAPKKK Raf, which in turn phosphorylates the MAPKK Mek1/2, which finally phosphorylates the MAPK Erk, leading to its activation [24,25]]. p38 MAPK can also be activa ted by tyrosine receptor kinases . In human vein endothelial cells (HUVEC), vascular endothelial growth factor (VEGF) activates both Erk and p38 MAPK, while havi ng no effect on JNK . VEGF enhanced endothelial cell motility and supported actin rearra ngements . Further an alysis revealed that p38 activation was required for endothelial cell mo tility and actin rearrangement, while Erk was found to have no role in this process. . p38 MAPK The p38 MAPK family consists of , , , and isoforms . p38 MAPK can be activated by various growth factor s, cellular stress, bacterial bypr oducts and cytokines (Fig. 1-2) [28,29,30]. p38 MAPK activati on can be induced by G-protein c oupled receptors or tyrosine receptor kinases that activate MAPKKK. The MA PK pathway leading to p38 activation is well understood, with the exception of receptor activ ation of MAPKKK. It is thought that small GTPases may play some role in activating MAPKKK . Once recept or activation occurs, phosphorylation activates the MAPKKK MLK3 (m ixed lineage kinase) . MLK3 then phosphorylates the MAPKK Mek3/6, which in tu rn phophorylates p38 MAPK (Fig. 2) . Once p38 is phosphorylated on threonine 180 and ty rosine 182, it is now activated and can phosphorylate substrate proteins on serines or threonines . p38 activation causes increased
17 expression of mRNAs through stabi lization of transcripts or by an increased rate of transcription driven by specific promoters [34-37]. By phosphoryl ating transcription factors, p38 is able to increase the expression of speci fic transcripts . p38 is thou ght to stabilize transcripts by activating proteins that bind and protect mRNA [34,35,36] . p38 can also phosphorylate MAPKAP-2 (mitogen activated protein kinase activated protein), which can lead to Hsp27 activation and actin remodeling [38,39] Macrophage Apoptosis Much research has been conducted on th e effects of LT on ma crophages. Macrophages that have been sensitized by LPS (lipopolysaccharid es) or peptidoglycan are subject to activation induced apoptosis by LT . The exact Toll-like receptors (TLRs) that anthrax interacts with and the nature of the signals transmitted are not known [40,41]. What is known is that most TLRs activate the NF B (nuclear factor kappa B) and p38 MAPK pathways, which are involved in anti-apoptotic gene expressi on and transcriptional activation of specific cytokines [40,41]. LT alone is not sufficient to induce apoptosis. LT must be combined with TLR stimulation to induce apoptosis in human macrophages. This finding indicates that deat h signals are also transmitted via TLRs. Both NF B and p38 MAPK are thought to induce anti-apoptotic gene expression, thus counteracting the death signals sent from TLRs [40,41]. The current understanding indica tes that anthrax stimulates unknown TLRs, which results in both death signals and survival signa ls being sent from the TLRs [40,41]. NF B and p38 MAPK normally work together to inhibit the de ath signals, which result s in cell survival and normal immune function. Both p38 and NF B are thought to help pr otect cells from death signals by causing pro-survival gene expression. Anthrax bloc ks p38 MAPK activity by cleaving MAPKK 3/6, and NF B alone may not be sufficient in prom oting anti-apoptotic signals .
18 Lethal toxin can block the ability of dendritic cells to prime CD4+ T cells by down regulating co-stimulatory molecules, such as CD40 on dendritic cell surface . LT also blocks secretion of specific cytokines fr om dendritic cells including TNF(tumor necrosis factor alpha), IL-6, (interleukin-6) and IL -12 (interleukin-12) . It is thought that this reduction is mediated by disrupting MAPKK signaling. An inte resting difference between dendritic cells and macrophages is that dendritic cells are not prone to LT-mediated a poptosis . Recent work has also shown anthrax toxins preven t T-cell activation . This inac tivation is though t to occur as a result of p38 MAPK inactivation . The above pathways represent the first examples of how anthrax may inhibit the ad aptive immune response. Neutrophil Function Extensive data exist for toxin/macropha ge interactions, but little work has been performed with neutrophils. Neut rophils are an important com ponent of the innate immune system, being the first responders to arrive at the site of an infec tion . Neutrophi ls arise in the bone marrow from a pluripotent stem cell that upon division gives rise to another stem cell and a leukocyte progenitor, which ultimately give rise to white blood cells [45,46]. Two types of leukocyte progenitors ex ist: lymphoid progenitors, which pro duce both B-cells and T-cells, and myeloid progenitors, which will produce monoc ytes, eosinophils, basophils and neutrophils [45,46]. Immature neutrophils are called band cells, and these cells lack the ability to undergo chemotaxis, and possess relatively few granules. Ba nd cells are rarely seen circulating in the bloodstream of healthy adults but may appear duri ng times of infection [4 4]. Mature neutrophils are thought to undergo chemotaxis from the bone marrow to the blood stream. Blood plasma has been shown to have chemoattractive activity fo r mature neutrophils but has no effect on band cells .
19 The numbers of neutrophils increase dramati cally during times of infection. In the bone marrow there are relatively large stores of mature neutrophils that can be released upon infection . Up to 99% of the neutrophil population is stored in the bone marrow in healthy adults . Lipopolysaccharide and other bacterial com ponents are known to increase the number of neutrophils in circulation dr amatically through the induction of G-CSF (granulocyte colony stimulating factor) [48,49]. Along wi th increasing the numbers of circulating neutrophils, G-CSF also was found to increase the delivery of neutr ophils to sites of inf ection and increase the bactericidal activity of the cells in animal inf ection models [50,51]. Colo ny stimulating factors have also been used to treat severe neut ropenia in patients recei ving chemotherapy, with reasonable results being achieved . Mature neut rophils have a very short lifespan once they enter the circulation. The half-lif e of mature neutrophils is on ly 6h, and neutrophils make up 6070% of circulating white cells in healthy adu lts; thus it is very important for the myeloid precursors to produce many neutrophils . It is thought that progen itor cells have the ability to produce 1011 neutrophils per day in order to mainta in a blood count of 10,000 cells/l . Neutrophils also express TLRs that can function as microbial pattern recognition receptors . These receptors are able to r ecognize microbial compounds such as bacterial lipids (LPS and lipoproteins), nuc leic acids (DNA and double stra nded ribonucleic acids), and bacterial proteins such as fl agellin. Human neutrophils expres s all of the known TLR family members except for TLR-3 . Ac tivation of neutrophil TLRs re sults in increased rates of phagocytosis, priming of superoxide production, and synthesis of chemokines and cytokines . Neutrophils have a limited ability to produce certain chemokines and cytokines themselves . The amount of cy tokines produced by neutrophils is very low, as compared to
20 mononuclear cells. Proinflammatory molecules such as LPS, C5a (complement factor 5), and FMLP (formylated methionyl-leucyl-phenyalani ne) can stimulate neutrophils to produce chemokines and cytokines . Cytokines and chemokines can be produced by neutrophils in three main ways. Degranulation causes the release of cytokines from intr acellular stores within the neutrophil cytoplasm [56,57]. These cytokines are pre-made and stored in granules until stimulation by pro-inflammatory molecules such as IL-8 (interleukin-8) . Neutrophils can also synthesize certain cytoki nes, even though neutrophils themselves are terminally differentiated cells . IL-8 transcription is upregulated upon exposure to LPS [59,60]. Neutrophils also have the abil ity to cleave surface-bound inactive cytokines from their plasma membranes . Tumor necrosis factoris produced as a 26-kD prot ein that is cleaved to a 17kD active form by TNF-alpha-conve rting enzyme (TACE) . Chemokines are molecules that are able to at tract leukocytes to sites of infection or inflammation. Neutrophils have the ability to produce RANTES (regulated upon activation, normal T expressed and secreted), macrophage inflammatory protein-1 (MIP-1 ), MIP-1 , and MIP-1 [55,62,63]. These chemokines have the abili ty to attract monocytes, eosinophils, basophils and T-lymphocytes to inflamed areas [ 64]. Neutrophils also have the ability to produce IL-8, which attracts other neutroph ils to sites of infection. This process is referred to as auto amplification . Neutrophils also can produce certain cytoki nes in a limited manner. When compared to mononuclear cells, neutrophils are responsible fo r only 1.5% of the total cytokine array . Neutrophils have the ability to produce TNF, IL-1 (interleukin-1), IL -6 (interleukin-6), Il-12 (interleukin-12), interferon(INF), and interferon(INF) [65-68]. TNFis an important pro-inflammatory molecule that is responsib le for increased vasodilatation and increased
21 permeability of the vasculature . This res ponse allows neutrophils to transverse the endothelium so that an infection can be controlled . Interferoncauses the activation of macrophages to kill intracellula r bacteria that have been phagocytosed . Interleukin-8, C5a, and bacterial by-products such as FMLP attract neutrophils to sites of infection [71,72,73]. Neutrophils have both complement receptors and Fc receptors that can bind both complement-opsonized or IgG-opsonized bacteria, resulti ng in an increased rate of phagocytosis . Once phagocytosed, the bact eria are killed by an array of cytotoxic compounds such as proteases and reactive oxygen species. Neutrophils can also undergo degranulation to release proteases and reactive o xygen species into the surrounding tissues to kill pathogens . It is important that neutrophils are cleared fr om an infected area once the infection has been extinguished, because prolon ged exposure to the toxic compounds released by neutrophils can cause extensiv e tissue damage . In conc lusion, the primary role of neutrophils is to phagocytose and kill bacteria that have entered the human host, limiting the spread of the invading organism. To accomplish this feat, neutrophils have the ability to undergo chemotaxis to the site of infection. Circulating neutrophils generally roll along the endothelial surface by making weak contacts with E-selectin . The neutrophil ligand sialyl-Lewisx mediates the weak interaction with E-selectin on the endothelial surface. This interaction is not strong enough to hold neutrophils stationary agains t the blood flow in the vessel, so neutrophils roll along the endothelial surface, until they encounter a stimulat ory molecule such as IL-8 . Interleukin-8 causes the release of inte grins, such as LFA-1 (leukocyte func tional antigen-1) from intracellular stores, allowing neutrophils to make strong cont acts with the molecule ICAM-1 (intercellular adhesion molecules) on the surface of activated endothelial cells . Intercellular adhesion
22 molecule-1 expression is increased by pr o-infalmmatory molecules such as IL-1 , TNF , and LPS . This tight binding allows the neut rophil to become attached to the vascular endothelium at sites where tissue damage or in fection are occurring. The inflammatory response causes increased permeability of the vasculat ure allowing neutrophil diapedesis into the surrounding tissue. Once the neutro phil enters the tissue space, it can follow the chemoattractant signal to challenge any invading microorganisms. Neutrophil Chemotaxis Chemotaxis requires rapid assembly and disass embly of actin filaments. The assembly of F-actin is controlled by an extensive cell-signaling cascade. FMLP stimulates chemotaxis through a heterotrimeric G-protein coupled recep tor pathway (Fig. 1-3) . This receptor recognizes formylatyed peptides at the aminoterimnus, and neutrophils are thought to have about 50,000 of these per cell  . Bacteria produce mRNA that are formylated on their aminoterminus, whereas human mRNAs are not formylat ed at the amino terminus. This allows for neutrophils to become active in response to bacterial mRNA but not human mRNA . FMLP is a short formylated peptide th at will stimulate neutrophil ch emotaxis and can be used in motility studies. Once the G-protein receptor is activated, it causes the exchange of GDP for GTP in the subunit of the G-protein [ 81]. Once this occurs the subunit releases from the subunit, which results in two active G-proteins . The subunit activates phospholipase-C (PLC), which results in an oxidative burst . The subunit is responsible for chemotactic signaling . The subunit is thought to activate phosphoinsoitide-3 kinase (PI3K), which then can activate guanosine exchange factors (GEF s) that cause the release of GDP for GTP in small GTPases such as Rac, Rho, and Cdc 42 [84,85,86]. The small GTPases are thought to activate N-WASP (Wiscott -Aldrich syndrome protein), which results in Arp2/3 (actin related
23 proteins) activation [87,88]. Arp2/3 can then nucleate new actin assembly. It is important to note that most reviews of actin assembly focus on G-proteins and phosphoinositides as key signaling molecules for actin assembly. MAPK pathways have been neglected by members of the actin community and have not been extensively studi ed as potential actin re gulatory pathways. Actin Structure and Function Actin is a globular 43-kD pr otein that exists in two fo rms: monomeric (G-actin) and filamentous (F-actin). The assembly of actin monomers into filaments provides the force necessary for cell motility. Actin filaments are polar structures that have two distinct ends: the plus end (more dynamic) and the pointed end (les s dynamic) . The plus end is also referred to as the barbed end of the actin filament. Mono mers can add or leave the barbed end at a much faster rate than the pointed end of the filame nt . Actin polymerizat ion has three different phases (Fig.1-4): nucleation, elonga tion and steady state . The nucleation stage is the ratelimiting step in actin assembly and requires the bi nding of three actin monomers to initiate actin assembly. The elongation phase is the period of rapid addition of monomers to the filament. The steady state phase occurs when the addition of actin monomers onto filaments is equal to the disassembly of monomers from the filaments. Th e concentration of monomers left in solution during steady state is referred to as the critical concentration. Th e critical concentration at the barbed end of the actin filament is less (0.1M) , while that of the pointed end of the actin filament is 0.6M . This means that the affi nity of the barbed end for actin monomers is greater than that at the pointed end of the f ilament. Thus, if the mono mer concentration is < 0.6M, monomers will only add to the barbed en d of the filament, until 0.1M is reached, and will simultaneously loose monomers off of the pointed end because the monomer concentration is below that of the pointed end critical. This phenomeon is referred to as treadmilling and results in no net growth of the actin filament.
24 Actin polymerization requires ATP (adenosin e triphosphate) in order for assembly to occur . Actin is an ATPase that can c onvert ATP into ADP (adenosine diphosphate) and inorganic phosphate. The ATPase activity of monom eric actin is much less than that of the monomers once they are incorporated into fila ments . Typically, ATP-actin monomers are added to the barbed ends of actin filaments by the monomer sequestering pr otein profilin (Fig.15) . Profilin binds ATP-actin and allows add ition of the actin monomers only at the barbed end of the filament . Once the monomer is a dded to the filament, ATP is hydrolysed to ADP. Adenosine diphosphate-actin is then released from the filament at the pointed end . Once this occurs, profilin binds to ADP-actin and helps ex change the ADP molecule for an ATP molecule so that the cycle can repeat itself . Actin Binding Proteins There are several classes of acin-binding proteins that ar e required to control actin assembly. Monomer sequestering proteins such as profilin and thymosin 4 (T 4) have the ability to bind monomeri c actin. [93,94]. Thymosin 4 is primarily responsible for preventing non-specific nucleation by blocking binding of both the barbed and pointed ends of the actin monomers . The primary role of profilin is to deliver actin monomers to the barbed end of growing filaments [93,96]. Profilin can take monomers from T 4, thus allowing addition of new monomers to the barbed ends of actin filaments [93,96]. As stated earlier, profilin can also aid the exchange of ADP for ATP in actin monomers . Capping proteins, such as gelsolin, CapG and CapZ can bind to and block (cap) the barbed ends of actin filaments . In fact, most filaments in the cell are capped, and this prevents unwanted filament growth. Once the barbed end is capped, monomers cannot add nor can they disassemble from the filament. Capping proteins therefore, help to maintain the
25 preexisting actin filaments in a cell. CapG is sensitive to both calcium levels in the cell and phosphoinositide 4,5 bisphosphate (PI(4,5)P2) . Calcium is needed for CapG to cap the barbed ends of filaments, and PI(4,5)P2 can cause filaments to be uncapped. Severing proteins such as ge lsolin, can bind to the sides of preexisting filaments and sever them . Gelsolin also can cap the barb ed end of the filament that it just severed. Severing is important for remodeling of the actin cytoskeleton, and also for actin assembly, because severing actin filaments followed by uncapping creates new barbed ends. Intracellular calcium levels and PIP(4,5)P2 control gelsolin activity in a similar manner to CapG [99,100]. As mentioned earlier, the rate-limiting step for actin assembly is nucleation. To bypass this thermodynamically unfavorable process, cells ha ve proteins that serve as nucleation factors. The most well known is the Arp2/3 complex, that can bind to 3 actin monomers so that assembly can be initiated . The Arp2/ 3 complex binds to the pointed ends of the monomers, thus allowing growth from the dynamic barbed end of the filament . The Arp2/3 complex is activated by small GTPases such as Rac, Rho, a nd Cdc42 . The activation of different Rho family GTPases has been shown to result in diffe rent types of actin st ructures. Rho activation causes the formation of stress fibers and focal contacts . Th e activation of Rac causes the formation of large lamellipodia, and the activat ion of Cdc42 causes the formation of filopodia [104,105]. These small GTPases, once activated, bi nd and activate WASP, which then binds to the Arp2/3 complex . Once this occurs, the Arp2/3 complex underg oes a conformational change that allows the monomers to bind, whic h results in nucleation of actin filaments. Listeria mononcytogenes The gram-positive intracellular pathogen L. monocytogenes has the ability to induce actin assembly once inside the cytoplasm of a host cel l . Listeria accomplishes this feat by taking over the host cell actin machinery and uses these proteins to pr opel itself through the cytoplasm
26 into adjacent cells (Fig.1-6) . By doing this, Listeria avoids host defenses such as complement and antibodies. Listeria has two important proteins th at are required for entry into non-phagocytic cells. Internalin A interacts with E-cadherin, and internalin B interacts with gC1qR and the hepatocyte growth factor receptor Met on the cell surface of the host to gain entry inside the cell [108,109]. In ternalin A entry require s the activation of both and catenins . The catenins cause rearrangement of the actin cytoskeleton result ing in phagocytosis of Listeria . The exact mechanism of how Listeria causes actin rearrangement s via catenins is not well understood, and more research is needed to decipher the pathway . It is known that catenins are able to join E-cadherin to the actin cytoskeleton, and this is required for Listeria entry . Internalin B lead s to the activation of phsophoinositide 3 kinase (PI3K), which produces phosphoinositide 3,4,5 triphosphate (PI(3,4,5)P3) and activates the Erk MAPK pathway [111,112]. Phosphpinositide 3,4,5 triphospha te production and Erk activation promote actin rearrangement and internalization of extracellular Listeria [111,112]. Actin rearrangement is promoted by the recruitment a nd activation of the Arp2/3 co mplex. Arp2/3 is activated by WASP upon internalin B binding to the Met receptor . WASP activation itself is dependant on small GTPases Rac and Cdc42 . These sm all GTPases become active after internalin binding. Once inside the cell, Listeria is contained within a phagosome and must escape to prevent killing by lysosomal components [22,23]. Listeria produces listeriolysin O, an exotoxin that lyses the phagosome so that Listeria can escape into the cytoplasm . The Listeria protein ActA is sufficient to produce actin asse mbly . ActA mimics the host protein WASP, activating the Arp2/3 complex to nucleate actin assembly (Fig. 6) [106,116]. ActA also recruits VASP (vasodilator-stimulated phosphopr otein) to the actin tails of Listeria . VASP can bind to proflin:actin and provide a source of monomers for new actin assembly . In extracts,
27 Listeria actin-based motility requires very few host proteins to generate movement. Arp2/3 is required to nucleate actin assembly, actin de polymerizing factor (ADF) helps to recycle monomers from the filaments, and capping proteins such as CapZ that block the barbed ends of actin filaments . This motility can be enha nced by the addition of profilin, VASP and the actin bundling protein -actinin . LT Paralyzes Neutrophil Motility by Blocking Hsp27 Phosphorylation I have demonstrated that LT can block neut rophil actin assembly and reduce the speeds of neutrophils stimulated with FMLP. Neutr ophil actin assembly was also reduced in the presence of a p38 MAPK inhi bitor SB203580. Lethal toxin and SB203580 reduced intracellular Listeria monocytogenes velocities while shortening actin rock et tails. The reduction in neutrophil chemotaxis/chemokinesis was accompanied with a lack of phosphorylated Hsp27 (heat shock protein 27). Hsp27 was found to be a potent actin monomer sequestering prot ein that can inhibit actin polymerization when Hsp27 is not phosphory lated. Phosphorylation reversed this effect. Non-phosphorylated Hsp2 localized to the lead ing edge of neutrophil lamellipodia and Listeria actin tails. Phosphorylation also reversed the localization, indicating th e potential for Hsp27 to have different functions, depe nding on phosphorylation status. I hypothesize that anthrax LT blocks actin assembly by increasing the intr acellular concentration of non-phosphorylated Hsp27, thereby providing more monomer sequest ering protein, which reduces the amount of available monomers for actin assembly. I showed that p38 MAPK may be an important signaling molecule for the assembly of new actin filame nts, highlighting previ ously unappreciated signal transduction pathways for actin-based motility.
28 Figure 1-1. Mitogen activated prot ein kinase (MAPK) signaling pa thways. There are 3 important MAPK: p38 MAPK, which is stimulated by st ress and cytokines; JNK MAPK, which is activated by cellular stress; and the Erk MAPK pathway, which is activated by mitogens. Receptor activation leads to a phosphorylation cascade that begins with MAPKKK activation and ends with a particular MAPK be ing activated (Erk, JNK or p38). The MAPK can then phosphorylate substr ate proteins on serine or threonine residues, thus activating the effector protein. Roux PP, Blenis J. ERK and p38 MAPK-activated protein kinase s: a family of protein kina ses with divers e biological functions. Microbiol Mol Biol Rev 2004; 68:320-44. Reproduced with permission from ASM press and authors.
29 Figure 1-2. The p38 MAPK pathway is activated by pro-inflammatory cytokines and bacterial byproducts such as FMLP. Receptor activa tion leads to phosphorylation of the MAPKKK MLK3, that phophoryles the MAPKK Mek3/6, which in turn phosphorylates p38 MAPK. Once p38 is activ ated it can phosphorylate effector molecules such as MAPKAP-2/3 that can promote actin rearrangement. Huang C, Jacobson K and Schaller MD. MAP ki nases and cell migration. J Cell Sci 2004;117:4619-28. Reproduced with permissi on of The Company of Biologistsâ€™.
30 Figure 1-3. The neutrophil FMLP receptor signa ling pathway. G-protei n coupled receptors activate G-proteins which then have the ab ility to activate guanosine exchange factors (GEFs). GEFs then cause the activation of small GTPaseâ€™s such as Cdc42, Rac, and Rho. Cdc42 can then activate N-Wasp, whic h leads to the activation of Arp2/3,which can now nucleate new actin assembly. Phos phoinositdes are thought to uncap gelsolin from actin filaments leading to barbed e nd growth. Finally, p38 activation leads to phosphorylation of Hsp27 causing the release of actin monomers. The p38 pathway has been overlooked by cell motility reviews. Cicchetti G, Allen PG and Glogauer M. Chemotactic signaling pathways in neutrophi ls: from receptor to actin assembly. Crit Rev Oral Biol Med 2002;13:220-8. Reprodu ced with permission from Critical Reviews in Oral Biology & Medicine. MLK3 MAPKK3/6 p38 MAPK MAPKAP2/3 Hs p 27
31 Figure 1-4. The growth curve for actin assembly. The nucleation stage (lag) is the rate-limiting step that requires three actin monomers to come together in-order for actin assembly to begin. The elongation stag e is intially lin ear and depends on the number of filament ends and the actin monomer con centration. Steady state occurs when the amount of monomers being added to the f ilament equals the amount of monomers leaving the filament, resulting in no net growth. The amount of actin monomers in solution during steady state is called the critical concentration. Lag Elongation Steady State Actin polymerization Time
32 Figure 1-5. Actin filament treadmilling. ATP-actin monomers are usually added at the barbed end of an actin filament. Once the monomer becomes incorporated into the filament, the ATPase activity of the monomer cau ses the hydrolysis of ATP to ADP + inorganic phosphate. Eventually the ADP-actin monomer is released from the pointed end of the actin filament. Treadmilling occurs when the concentration of monomers in solution is greater than the critical con centraion of the barbed end, but is less than that of the pointed end. Profilin binds to th e monomers and catalyzes the exchange of ADP for ATP so that the cycle can repeat itself. barbed end pointed end Profilin ADP-Actin ATP-Actin ATP ADP ADP ADP
33 Figure 1-6 . L. monocytogenes uses the host cells actin machiner y. ActA recruits and activates the Arp2/3 complex to nucleate actin a ssembly, by mimicking the Wiscott-Aldrich syndrome protein (WASP). ActA also recruits the vas odilator-stimulated phosphoprotein (VASP) to Listeria actin tails. VASP is then able to supply proflin:actin to the polymerization zone. Capping pr oteins are required to produce direcrtional movement. Bund ling proteins such as -actinin stablize existing actin filaments. Reprinted from Cossart P, Bierne H. The use of host cell machinery in the pathogenesis of Listeria monocytogenes. Curr Opin Immunol 2001;13:96-103 with permission from Elsevier.
34 CHAPTER 2 MATERIALS AND METHODS Neutrophil Motility and Actin Assembly Toxin Purification Dr. Conrad Quinn of the CDC in Atlant a Georgia kindly provided purified toxin components. Toxin components were purified as described previously . Briefly, culture media were filtered through a 0.22 m filter, followed by diethylaminoethyl cellulose (DEAE) anion exchange chromatography. The resulting t oxin components were then subjected to gel filtration and hydrophobic interaction fast protein liquid chromatography (FPLC) as previously described . Cultures of 15 liters were found to generally yi eld 8mg of PA, 13 mg of LF, and 8 mg of EF with purity assessed by coomassie blue staining to be 90% . Neutrophil Isolation, Toxin Trea tment and MAPKK Western Blot Whole blood was collected from healthy vol unteers as describe d previously . Informed consent was obtained from all subj ects, and the study followed US Department of Health and Human Services guid elines. Protocols were also approved by the University of Florida Institutional Review Bo ard (IRB). Neutrophils were isol ated from whole blood by using Ficoll/Hypaque gradient medi a (ICN Biomedical, Irvine, CA ) followed by hypotonic lysis of contaminating red cells. The re sulting neutrophil populat ion was determined to be 99% pure as determined by microscopic examination. Cells we re suspended in RPMI without L-glutamine (Mediatech Inc., Herndon, VA) at a concentration of 1.0 x 106 cells/ml unless otherwise noted. 50ng/ml of LT was added to the cells for 2h at 370 C while gently rotating to prevent cell clumping, unless otherwise noted. Control cells were treated with buffer al one and incubated as above. Western blotting was performed by usi ng anti-N-terminal MA PKK-1 (Upstate Cell
35 Signaling, Lake Placid, NY), whic h recognizes the N-terminal 7 amino acids. To prevent nonspecific proteolysis, neutrophils were pretreat ed with diisopropyl-fluorophosphate (DFP) as previously described, be fore blotting . Neutrophil Annexin V Staining, Propid ium Iodide Staining and NBT Test Annexin -V staining was perf ormed on neutrophils using th e Annexin-V-FLUOS staining kit (Rochelabs, Penzberg Germany) by following the manufactureâ€™s protocol. Stained cells were subjected to flow cytometry (FACScan, BD Bi osciences, San Jose, CA) using an excitation wavelength of 488nm and emission spectrum of 5 18nm to detect annexin-V stained cells and 617nm to detect propidium iodide -stained cells. As a positive control, I us ed mouse anti-human Fas IgM to stimulate neutrophil apoptosis. Fas-me diated apoptosis was carried out as previously described . Briefly, neutroph ils were incubated with mouse anti-human Fas IgM (500ng/ml) for 2h or nonspecific IgG (500ng/ml) before flow analysis. In t oxin-treated cells, anti-Fas IgM was added to the cells at the same time as LT (50ng/ml) before flow cytometry analysis. To determine if LT resulted in the block of s uperoxide production, I performed the nitroblue tetrazolium test. The nitroblue tetrazolium test (NBT) was performed by following the manufactureâ€™s protocol (Sigma). Briefly, ne utrophils were stimula ted by 200ng/ml of phorbol myristate acetate (PMA), and formazan depos ition was analyzed by light microscopy. One hundred cells were counted for each condition. Neutrophil Chemokinesis, Chemotaxis, and Polarization Neutrophils (1.0 x105) were plated in a final volume of 2ml RPMI without L-glutamine on glass bottom microwell dishes (Matek Cu ltureware, Ashland, MA ) coated with 0.1% fibronectin (Sigma, St. Louis MO ) after treatment with buffer or toxin as described above. Neutrophil chemokinesis was measured by dispensing a final c oncentration of 1 M of FMLP
36 (Sigma, St. Louis, MO) directly into the microwell dish. Time-lapse phase contrast images were captured at 20s intervals using an inverted mi croscope with a 100X lens (Nikon, Tokyo, Japan) and a cooled charge-coupled device camera (Hamamatsu, model C5985). Images were processed using Metamorph 4.0 image software (Universal Imaging, West Chester, PA). Neutrophil velocities were determined by us ing the Metamorph program as pr eviously described . The percentage of polarized neutrophils (having a distinct lamellipod or uropod), was determined after a 4-12 minute exposure to FMLP by a blin ded observer. Chemotaxis was measured by using a FemtoJet needle (0.5 m tip diameter, Eppendorf, Hamburg, Germany) containing 10 M FMLP. The needle was introduced at one corner of the coverslip and dispensed at 15 psi using an Eppendorf micromanipulator. Velocities were determined by using the Metamorph program . Neutrophil Phalloidin Staining Following toxin or mock treatment, 1.0 x106 neutrophils were stimulated with 1 m FMLP for 0, 5,10,15,30,60, and 120 seconds followed by 3.7% formalin fixation as previously described . Neutrophils were then pe rmeabilized with 0.2% Triton-X100 solution containing phalloidin conjugate d to Alexa-488 (Molecular Probes, Eugene, OR) . FACS analysis was carried out using an excitation wavelength of 488nm and emission wavelength of 518nm. The MAPKK-1 inhibitor PD98059 and the p38 MAPK inhibitor SB203580 (Calbiochem, Darmstadt Germany) were used as previously described . Neutrophils were treated with 100 M of inhibitor at 370 C for 10 min prior to FMLP stimulation. Triton X-100 Insoluble Cytoskeleton Neutrophils were isolated fr om human volunteers by dextran sedimentation as in  and were adjusted to 1.0 x 107 cells/ml. The triton X-100 inso luble cytoskeleton assay was
37 performed exactly as previously described [ 125]. Briefly, cells were stimulated with 1 M FMLP for 0, 20, or 40 seconds and the reaction was st opped by the addition of 1:1 volume of a 2% triton X-100 stop solution (2% triton, 160mM KCl, 40mM imid azole HCl, 20mM EGTA, and 8mM sodium azide pH 7.0) containing complete mini protease inhibitor (Roche). Cells were centrifuged, and the resulting pelle t was subjected to sodium dod ecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The density of the 43-kD actin band was quantified by using Alphaimager v5.5 (Alpha Innot ech, San Leandro CA). LT and Hsp27 2D SDS-PAGE and Hsp27 Western Blotting 2D (two dimensional) SDS-P AGE was carried out exactly as described previously . HeLa cells were treated with 1 g/ml LT overnight or neutrophils were treated with 500ng/ml for 2h before being stimulated with 1M FMLP for 30s. Cells were lysed with non-denaturing lysis buffer (1% triton X-100, 50mM Tris-Cl pH7.4, 150 mM KCl, 5mM EDTA, 0.02% sodium azide, 2mM imidazole, 1mM sodium fluoride, and 1mM sodium orthovadanate) containing complete mini protease inhibitor (Roche). Cell lysates were sonicated 3 times for 30s followed by the addition of 4 volumes of acetone. Protein pr ecipitation was carried out for 4h at -200 C followed by centrifugation at 800 x g. The resulting protei n pellet was air dried and resuspended in rehydration buffer (8M Urea, 2% CHAPS, 0.5% IPG buffer 3-11, trace of bromophenol blue, and 2.8mg/ml DTT). Immobiline DryStrip (A mersham Pharmacia Biotech) pH 3-11, 18cm isoelectric focusing strips were rehydrated with 600 g of protein as determined by Bradford protein estimation for 12h followed by isoel ectric focusing for 6h (500V for 1h, 1000V for 1h, and 8000V for 4h) using the Ettan IPGphor is oelectric focusing cell (Amersham Pharmacia Biotech). After focusing, the strips were equ ilibrated in SDS equilibration buffer (50mM Tris-Cl
38 pH 8.8, 6M Urea, 30% glycerol, 2% SDS, tr ace of bromophenol blue and 10mg/ml DTT) followed by second dimension resolution on 12.5% polya crylamide gels. Gels were silver stained using Silver Quest (Invitrogen, Carlsbad, CA) by following the manufactur eâ€™s protocol. Protein spots were excised from the gel and subj ected to matrix assisted-assisted laser desorption/ionization-time of flight (MALDI-TOF). Protein identification was carried out at the University of Florida Interdisciplinary Center for Biotechnology Research (ICBR) core facilities. Hsp27 western blots were performed on 2-dime nsional separated proteins by transferring proteins to nitrocellulo se membranes (BioRad) followed by detection with total anti-Hsp27 and anti-phospho Ser82 Hsp27 antibodies (Cell Si gnaling, Beverly, MA). Primary anti-Hsp27 antibodies were used at 1:1000, and secondary anti bodies were used at 1:2000 dilutions. Western blots were also performed using 1-dimensiona l SDS-PAGE alone. HeLa extract (50g) was loaded onto 10% polyacrylamide gels, and both anti-total Hsp27 and phospho-Ser82 Hsp27 antibodies were used to probe the blot as desc ribed above. Western blots were also performed using both anti-phospho and anti-total p38 MAPK antibodies (Cell Signaling, Beverly MA). Super Signal West Pico Chemluminescent Substr ate was used for protein detection (Pierce, Rockford, IL). Listeria Infection and Phalloidin Staining Listeria infection and phalloidin st aining was carried out exactly as previously described . Listeria infections were allowed to proceed for 1h before the addition of 50ng/ml LT or 100M SB203580. Briefly, L. monocytogenes 10403S (1.0 x106 cells/ml) were used to infect 1.0x105 HeLa cells, followed by the addition of 50 g/ml gentamicin sulfate 60 min after infection to prevent extracellular growth. Listeria motility was observed 3-4h after infection by using phase microscopy as descri bed above. Velocity measurements were determined using the
39 Metamorph program . After Listeria infection was complete, cel ls were fixed with 3.7% formalin followed by treatment with 0.4% trit on containing Alexa-488 c onjugated to phalloidin (Molecular Probes, Eugene, OR) . Tail leng ths were determined by using the Metamorph program. Recombinant Hsp27 Recombinant Hsp27 was kindly supplied by Dr. J acques Landry at the Universit Laval, Qubec, CA. Wild-type and phosphorylation mutant Hsp27 proteins were described previously . Site directed mutagensis was performed on wild-type hamster Hsp27 to mutate serines 15 and 90 to alanines (pseudo-non phosphorylated mutant) and to glutamic acids (pseudophosphorylated mutant). Pseudeonon phosphorylated mutant protei n is referred to as Hsp27 AA and pseudo-phosphorylated mutant prot ein is referred to as Hsp27 EE. Listeria Motility in Rat Brain Extracts Rat brain extracts (RBE) were prepared and motility assays were performed exactly as previously described . Brie fly, rat brain was homogenized in equal volume extraction buffer (20mM HEPES, 100mM potassium acetate, 1mM magnesium acetate, 1mM EGTA, 100mM sucrose, 1mM ATP, 1mM PMSF, 1mM DTT, a nd Roche Complete Mini protease cocktail tables). Extracts were then subjected to sonica tion three times for one mi nute at a power setting of 5 and duty cycle of 80%. Samples were then subjected to centrifugation with a Beckman J221 centrifuge using a JA-10 rotor at 20,000 xg for 1h at 40C. The extracts were mixed with 1% v/v methyl cellulose and 1:6 energy mix (150 mM creatine phosphate, 20mM ATP, 2mM EGTA, 20mM MgCl2, and 2% sucrose). For motility assays 1 l iodoacetic acid-fixed Bug 666 was mixed with 1 l of 24 M rhodamine-actin, 6l of brain ex tract and various amounts of Hsp27. The resulting mixture (2l) wa s placed on 1% w/v BSA-coated slides sandwiched between
40 22x22mm coverslides. The slides were allowed to sit at room temperature for 30min before viewing, using the inverted microscope as descri bed above. Tail lengths were measured using the Metamorph 4.0 program as described above. RNA Interference HeLa cells were split into 35mm glass botto m dishes on the day before transfection, so that cells were 50-60% confluent at the time of RNAi (ribonucleotide in ferference) transfection. Transfection of Hela cells was performed w ith Lipofectamine 2000 (Invitrogen, Carlsbad, CA) by following manufactures protocol . Briefly, 5l of Lipofectamine was resuspended in 500l of plain DMEM. 100nM of both control and Hsp27 si RNA were resuspended in 500l of DMEM and were allowed to set a room temperature for 5min. The siRNA was combined with the Lipofectamine for 20min before evenly dispensi ng onto the cells. Media were changed 6h after transfection to limit cell death. Hsp27 siR NA UGAGACUGCCGCC AAGUAA (target sequence) and control siRNA (UAAGGCU AUGAGAUAC) were purchased from Dharmacon. (Chicago, IL). siRNA was prepared according to manufactures protocol. After 48h, Listeria infections were performed as described above. Western blots were also performed using anti-Hsp27 antibodies to confirm knock-down. Neutrophil and Listeria Immunoflorescence HeLa cells were infected with Listeria as described above. Anti-Hsp27 antibodies (Cell Signaling, Beverly, MA) were used to stain both total Hsp27 and phosphorylated serine 82 isoforms in Listeria infected HeLa cells. Cells were fi xed with 3.7% formaldehyde followed by 0.2% triton X-100 permeabiliz ation. Cells were blocked with 4% fetal bovine serum followed by the addition of anti-Hsp27 (1:100). Rhodamine c onjugated to anti-rabbit IgG or anti-mouse IgG
41 was used for secondary antibody (1:1000) detect ion. Neutrophils were isolated and were prepared for staining as described above for Li stera-infected HeLa cells. Both anti-Hsp27 and anti-phosphorylated serine 82 were used to st ain neutrophils after 1M FMLP was added for 5min. Confocal Immunofluorescence of Neutrophils Human neutrophils were adhere d to fibronectin-coated cove rslips and stimulated with 1M FMLP for 5 minutes. Cells were fixed a nd stained with Hsp27 an tibodies (Cell Signaling, Beverly MA). Confocal images were collect ed every 0.5m. Two cr oss-sections were superimposed to show Hsp27 localization wit hout thickness distortion. One section at the adherent surface (pseudocolor green image), and the s econd section was taken 3m higher (pseudocolor red image). The two images were then overlaid. Actin Kinetics Monomer Sequestering Monomer sequestering assays were performed as described previously . Briefly, 2 M gelsolin-actin (1:20) filaments were polym erized in S2 buffer buffer (0.5mM ATP, 1mM dithiothreitol, 1mM CaCl2, 0.1M KCl, 1mM MgCl2 and 10mM imidiazole HCl, pH 7.5). Various concentrations of Hsp27 were prei ncubated with 1M G-actin for 3min. 120nM gelsolin-actin filaments were then added to th e Hsp27 and G-actin solution to initiate actin assembly in 2XP buffer (20mM imidazole, 0.2M KCl, 4mM MgCl2, 2mM ATP, and 2mM DTT). Fluorescence intensity was measured as a function of time. Rates were plotted as the change in intensity/time during the linear growth phase. Fo r time course experiments, Hsp27 (13.5g/ml) was pre-incubated with 1M G-actin for 0, 1, 3, a nd 5min before being added to the gelsolin seeds.
42 Barbed End Assembly With F-Actin Seeds N(1-Pyrenyl)iodoacetamide (pyrene) labeled actin was prepared from rabbit skeletal muscle as described previously . Pyrene was conjugated to actin on cysteine residue 373. Briefly, F-actin was added to pyrene at a 1:1 mo lar ratio in the presen ce of 0.1 M KCl, 1 mM MgCl2, 0.1 mM CaCl2, 0.2 mM ATP, 1 mM biocarbonate (pH 7.6) and 1 mM sodium azide. This was incubated for 20 h in darkness, before adding Whatman CF-11 cellulose to absorb unreacted dye. After dye removal, the labeled F-actin was pelleted by centrifugation and resuspended in 0.2 mM ATP, 0.1 mM CaCl2, 2 mM imidazole-HCl (pH 7.0), 1 mM 2-mercaptoethanol, and 1mM sodium azide. G-actin was then further purified by gel chromatography on Sephacryl S-200 with elution by G-buffer . Actin fluorescence ha s been shown to increase 25 fold upon actin filament formation, thus actin assembly can be measured by fluorescence intensity . F-actin formation is thought to cause a conformational change in the ac tin monomers near cysteine 373, resulting in increased fluorescen ce. . The critical concentra tion of the pyrene labeled actin was determined to be 0.08 M. Spectrin 4.1 was prepared from erythrocyte ghosts as described previously . Spectrin 4.1 is a component of the erythroc yte plasma membrane. This complex consists of both and spectrin, band 4.1, ankyrin, ba nd 3, glycophorin, and actin . The spectrin 4.1 complex, gives structural integrity to the erythr ocyte outer membrane. Purified spectrin 4.1 can be used to nucleate barb ed end actin assembly because it contains actin filaments with capped pointed ends and free barbed ends. . It is also thought that spectrin 4.1 may stabilize monomers that are trying to nucleate . For the barbed end assembly assays, 16g/ml of spectrin 4.1 was added to 2 M G-actin and was polymerized in S2 buffer. The resulting spectrin-actin filaments were then diluted to 120nM in a glass cuvette containing various concentrations of Hsp27, 1 M G-actin, and 200 l of 2XP buffer (20mM imidazole HCl,
43 .2M KCl, 4mM MgCl2, 2mM ATP and 2mM DTT) in a final volume of 400 l. Polymerization was assessed by measuring the fluorescence intens ity with a fluorimeter (Fluorolog-3 S.A. Instruments, Edison, NJ). The change in actin fluorescence was measured at 407nm after excitation at 365nm . Barbed End Actin Assembly With Spectrin 4.1 Seeds To determine if Hsp27 inhibited actin assembly by barbed end capping or monomer sequestration, I performed experiments in whic h Hsp27 was pre-incubated with spectrin 4.1 for 3min. In the case of incubati on with spectrin 4.1, either 5.4, 13.5, or 27g/ml of Hsp27 was incubated with spectrin 4.1 (6.5g) for 3min prior to the addition of 1M Gactin to initiate actin assembly. All polymerizations we re carried out in S2 buffer. Critical Concentration Assays Actin filaments (2M) were polymerized in S2 buffer overnight. Different amounts of Gactin (0.5M, 0.8M, 1M and 1.5M) were pre-in cubated with either 0, 50 or 100g/ml Hsp27 (final concentration) for 3 mi nutes. Actin filaments (120nM) we re then added along with 2XP buffer to initiate actin assembly. Rates were plotted as the change in intensity/time during exponential growth phase Disassembly Assays Disassembly assays were performed as described previously . Briefly, 2 M spectrin-actin filaments were polymerized in S2 buffer. F-actin was sheared mechanically with extended length pipet tips (Fisher Scientific, Su wanee, GA) by pipetting up and down four times in a glass cuvette. The sheared filaments were then diluted to 100nM by adding S2 buffer and various concentrations capping pr otein. Depolymerization was m easured as a function of time using the fluorimeter as described above.
44 Monomer Sequestration After Treatment of Hsp27 With MAPKAP-2 Hsp27 (5.4g/ml or 13.5g/ml) was incubate d with 2.5 g/ml of active recombinant MAPKAP-2 (Upstate Cell Signali ng, Lake Placid NY) in kinase buffer (20mM HEPES, 10mM MgCl2, 10mM MnCl2 and 1M ATP) for 30min. After phosphor ylation was carried out, 1M of G-actin was pre-incubated with Hsp27 for 3m in. To initiate actin assembly, 120nM of gelsolin:actin seeds (1:20) wa s added to the G-actin. Recombin ant gelsolin was produced as described previously . To verify phosphorylation of Hsp27 by MAPKAP-2, phosphoserine 82 Hsp27 antibodies were used to pe rform western blots on samples that were phosphorylated for 30 min. Invitro phosphorylation was carried out in kinase buffer using 0.67g of Hsp27 and 0.1g of active MAPKAP-2. Statistical Analysis The Wilcoxon two-tailed nonparametr ic test and the Fischerâ€™s exact test were used to determine statistical significance. In all experi ments except for FACS analysis, n refers to the number of cells analyzed. In most experiment s a minimum of 3 separate experiments were performed, unless noted.
45 CHAPTER 3 LITERATURE REVIEW Anthrax Clinical Data A successful bacterial infection must clearly be able to thwart the neutrophil response to establish significant numbers of bacteria in the host. One possible mechanism in which this could be accomplished is by blocking the neutrophil response by inhibiting chemotaxis. For chemotaxis to occur, new actin assembly must take place. Actin assembly can result from the nucleation of new filaments by the Arp2/3 comple x, uncapping of old filaments, and finally by severing proteins such as gelsolin that could create new barbed ends that could accommodate new monomers. B. anthracis toxins could potentially act to li mit chemotaxis by targeting any of the aforementioned mechanisms by which the cell creates new actin filaments. Clues supporting the possibility of a neutrophil chemotatic defect are provided by the firs t ten inhalation anthrax cases of 2001 . All patients pr esented with near normal or s lightly elevated white blood cell counts . The differential wh ite blood cell count showed that neutrophil levels were within normal limits or slightly elevated upon initial hospital presentation . This could mean that neutrophil migration from the bone marrow to the peripheral blood stream was defective or that neutrophil turnover was high once the neutroph ils reached the blood stream. Hemorrhagic pleural effusion fluid that was taken from patie nts was also noted to contain very few white blood cells, providing further eviden ce of poor chemotaxis into th e infected area . Lethal toxin could paralyze the innate immune re sponse by targeting macrophage apoptosis and neutrophil chemotaxis, which would render the pa tient dependant on an ad aptive response, which may still be days away. A poor innate immune re sponse may also be responsible for the severe bacteremia and sepsis that fo llows shortly af ter infection.
46 Previous Studies with LT and Neutrophils The effect of anthrax toxin on neutrophil ch emotaxis has not been revisited in nearly twenty years. It was shown that LF+PA, EF+PA, and EF+LF+PA enhance FMLP-directed chemotaxis, while not affecting random migration . Twenty years ago the target of LT was not known, and there was no way to determine the act ivity of the toxin preparation that was used. Additionally this previous study was performed without being able to determine the activity of the toxin. Today we have N-terminal specific antibodies that show e fficient cleavage of MAPKK, and we can judge the activity of LT. Th e researchers also incu bated the toxin for only 1h, and our findings indicate that longer incubation times are re quired for significant inhibition. When taken together, the clinical and research data call for a reevalua tion of anthrax toxinâ€™s effect on neutrophils. More recently, LT wa s shown to block FMLP-induced superoxide production in human neutrophils . High doses of LT were al so found to not significantly increase neutrophil cell death for over a seven hour period . Superoxide production was also inhibited by the p38 MAPK inhibitor SB 203580, indicating that LT blocks superoxide production by disrupting the p38 sign aling cascade. I have chosen to use LT to study the effects of anthrax on neutrophils because LF is a protea se that may actively cleave actin or other actinassociated proteins that may be involved in chemotaxis.
47 CHAPTER 4 RESULTS Anthrax Toxins Paralyze Neutrophil Actin Assembly Anthrax Lethal Toxin Does Not Cause Apoptosis or Necrosis in Human Neutrophils Lethal toxin has previously been reported to cause activation-induced apoptosis in macrophages , so I had to determine suitable LT concentrations that would not result in significant apoptosis within the 2 h time frame. We chose to use 2 h because this time frame has been reported to be sufficient for complete MA PKK clevage . To determine the extent of neutrophil apoptosis and necrosis after LT addi tion, I used Annexin-V and propidium iodide staining. Annexin-V specifically st ains apoptotic cells, while propi dium iodide stains necrotic cells, allowing us to decipher between the two events. As shown in Table 4-1, 50ng/ml of LT did not result in a significant amount of apoptos is or necrosis. Contro l cells exhibited 8.4% apoptosis and 4.9% necrosis, while LT-treated cells exhibited 8.2% apoptosis and 3.7% necrosis. For an internal control I used anti-Fas IgM to make sure our assay was functional. It has been previously shown that anti-Fa s IgM induces apoptosis in ne utrophils . Fas-mediated apoptosis without LT resulted in 19.3% apoptosis , while Fas+ LT-treate d cells exhibited 19.2%. As expected, I did not see an increase in necrosis in the presence of anti-Fas IgM. I also performed the nitroblue tetrazo lium test (NBT) to verify that superoxide production in neutrophils was not effected upon addition of LF+PA at 50ng/ml. The reduction of NBT by superoxide results in the deposition of blue fo rmazan crystals that can be seen by using light microscopy. Neutrophils were stimulated with 200ng/ml phorbol myristat e acetate (PMA) after incubation in buffer alone or 50ng/ml of LT for 2h. Unstimulated control cells exhibited 6.5% NBT positive cells, while unstimulated LT-treated cells resulted in 6% positive cells. Upon stimulation with PMA, 92% of control cells exhibited formazan deposition, while 90% of LT
48 treated cells resulted in deposition. This differe nce not significant (n= 100, p=1.0 Fisherâ€™s exact test). This finding shows that LT does not inhib it neutrophil superoxide production and that it is unlikely that LT is rendering neutrophils comple tely non-functional. To gauge the activity of 50ng/ml LT, I performed western blots with a N-terminal specific MAPKK-1 antibody. As shown in Fig. 4-1, complete cl eavage of MAPKK resulted within the 2h time frame with 50ng/ml LT. These results indica ted that 50ng/ml of LT does not result in significant apoptosis or necrosis within 2h, nor does it inhibit superoxide production, but does allow complete cleavage of MAPKK. Anthrax Lethal Toxin Impairs Neut rophil Chemotaxis and Chemokinesis I next examined whether 2h incubation w ith 50ng/ml LT interfered with neurophil chemokinesis. Neutrophils were incubated with 50ng/ml LT or buffer alone for 2h. Neutrophils were then allowed to attach to fibronectin -coated plates before stimulation with 1 M FMLP. Neutrophil velocity was recorded by time-lapse video microscopy as described in experimental methods. As shown in Fig. 4-2A, control neutro phils exhibited significan tly higher velocities than LT-treated neutrophils (Bars= standard error of the mean (S EM) of 11-14 neutrophils). This decrease in chemokinesis persisted for 12 min as shown in Fig. 4-2B (Bars = SEM of 135-210 measurements, n=14 cells/period; p< 0.0001). Neutrophils were also incubated with LT for only 1 h to help explain the differences between our observations and those in Wade et al . Neutrophils that were treate d for only 1h showed a significa nt chemokinesis defect (p< 0.001, 230-240 measurements; n= 30 cells), but velocity speed was reduced by only 33% as compared to 2h treated neutrophils, which e xhibited a 66% decrease in chem okinesis (Fig. 4-2 B). I also measured the ability of contro l and toxin-treated neutrophils to polarize after both 1h and 2h treatments. Neutrophils were stimulated with 1 M FMLP for 5 min before data were collected.
49 Phase contrast micrographs such as those s hown in Fig. 4-2 C and D showed that control neutrophils form broad lamellipodia and narro w uropodia (polarized morphology), while only one cell in the 2h LT treated micrograph formed a polarized structure. In the micrograph of treated cells most of the cells form non-polar ized rounded structures. These sample images reflected the overall differences in the two ne utrophil populations and were highly significant (Fig. 4-2 E; p< 0.0001). One-hour incubation with LT did not result in a significant difference in the ability of cells to polarize in response to FM LP, as compared to control cells (Fig. 4-2 E; p= 0.66). The ability of neutrophils to spread on an adherent surface after FMLP treatment was also assessed by measuring the area of individual cells. Ce lls that were treated with toxin for 2h had a mean area of 21 1.2 m2 (n=39 cells), while control cells had a mean area of 30 2.6 um2 indicating that toxin treatment interfered with spr eading (n=39; p= 0.013). In conclusion, 50ng of LT severely impaired neutrophil chemokinesis, po larization and spreading in response to FMLP. I next wanted to determine if exposure to 50ng/ml LT for 2h coul d inhibit neutrophil chemotaxis. Neutrophils were subjected to a chem oattractant gradient by constant infusion of a 10 M FMLP solution via a microinjection needle. As shown in Fig. 4-3 A, control neutrophils crawled toward the microinjection needle, wh ile LT-treated neutrophils failed to exhibit directional amoeboid movements. It was also of interest to note the polarized morphology of the control neutrophils, and lack of this morphology in cells treated by LT. Control neutrophils also exhibited higher velocities (0.090 0.003 m/sec, n=105-131), while treated neutrophils exhibited miniminal chemotaxis (0.004 0.001 m/sec, p< 0.001, n= 105-131; Fig. 4-3 B). I found that neutrophil chemotaxis (directional mo tility) was affected more than chemokinesis (random motility), indicating that LT may be able to disrupt ac tin signaling pathways that are required to detect chemo-attractant gradients.
50 Anthrax Lethal Toxin Blocks Neutrophil Actin Assembly Neutrophil actin assembly was assessed by using phalloidin conjuga ted to Alexa-488 as described previously [123,140]. Phalloidin is a well known mushroom toxin that specifically binds actin filaments. The relative fluorescence of cells stained with fluorescent conjugated phalloidin has been shown to be proportional to F-actin content [123,140]. Neutrophils were stimulated with 1 M FMLP for various time periods as shown in Fig. 4-4 A. After stimulation, neutrophils were fixed, permeab lized and stained with phalloidin as described in the experimental design. A peak in F-actin assembly occured at 30s for bot h control and LT treated cells (Fig. 4-4 A). However, the ex tent and the early time course of actin assembly were different in toxin-treated cells when compared to control cel ls (Fig. 4-4 A). In control cells F-actin content rose from 24 to a maximum relati ve F-actin content of 65 (3 fold change) fluorescent units;while toxin treated cells rose from 24 to 40 fluorescen t units (1.5 fold change; Fig. 4-4 A, SEM= 3 independent experiments). Also in the toxin treated cells, a d ecrease in F-actin content was observed at 5s, while a rise in F-actin conten t was observed in contro l cells (Fig. 4-4 A). Neutrophils that were treated with toxin for 1h did not exhibit a de fect in actin assembly, nor did they exhibit a delay in kinetics (Fig. 4-4 C, SEM 3 separate e xperiments). Neutrophils were also treated with various concentrations of LT fo r 2h (5ng/ml-500ng/ml) to see if a dose dependant block in actin assembly could be observed (Fig. 44 B). At 5s a small decrease in F-actin content occured with all three concentra tions of LT. A 50% reduction in F-actin content occured at 30s with all three concentrations of toxin. To veri fy my FACS data I performed the triton-insoluble cytoskeleton assay . This assay uses sedime ntation to differentiate longer actin filaments found in the triton-insoluble cy toskeleton (pellet) from short filaments and actin monomers (supernatant), and it is thought to be a more se nsitive assay than phalloidin staining. Cells were
51 treated with toxin for 2h be fore stimulation with 1 M FMLP at various time points. Cells were permeabilized, and F-actin was pelleted by centrif ugation as described in the experimental design. Densitometry scanning of th e resulting 43kD protein (actin ) showed that toxin treated cells exhibited a decrease in F-actin content, when compared to control cells (Fig. 4-4 D, mean of 2 experiments). Both of these experiments led me to believe th at low concentrations of LT can block F-actin assembly in reponse to the chemoa ttractant FMLP. Only a 50% reduction could be achieved by LT addition before FMLP stimulation. Th e fact that I could never completely block actin assembly led me to believe that there are multiple pathways by which FMLP can initiate actin assembly, and LT may be able to block only one of these pathways. p38 MAPK is Required For Neutrophil Actin Assembly in Response to FMLP Recent work has shown that p38 MAPK and Erk MA PK have distinct roles in regulating the actin cytoskeleton ]. Erk was required fo r chemokinesis, and p38 MAPK was required for chemotaxis, so we decided to investigate th e role of both Erk and p38 inhibitors on actin assembly . Since LT is known to cleave MAPK K 1-4 and 6/7, I wanted to see if the Erk inhibitor PD98059 could block neutrophil actin assembly in response to FMLP. PD98059 inhibits MAPKK1/2 preventing Erk activ ation. Cells were treated with 100 M of inhibitor 10min prior to stimulation with 1 M FMLP as described previously . As shown in Fig. 4-5 A, a slight increase in F-actin content occu red upon treatment with PD098059, as compared to control cells (2 separate experiments). This indicated that MAPKK-1/2 cleavage is not responsible for the reduction in neutrophil Factin assembly. However, the p38 inhibitor SB203580 blocked neutrophil actin assembly in res ponse to FMLP (Fig. 4-5 B). This indicated that MAPKK3/6 and the p38 MAPK pathway may be important for actin assembly in human
52 neutrophils. This pathway has not been previously recognized as an essential signal transduction pathway leading to the induc tion of actin assembly. LT Blocks Hsp27 Phosphorylation Anthrax Lethal Toxin and the p38 Inhibitor SB203580 Block Intracellular Listeria monocytogenes Actin Assembly To determine if anthrax lethal toxin could impair Listeria actin-based motility, I infected toxin treated HeLa cells (50ng/ml for 2h) with L. monocytogenes . As mentioned in the introduction, Listeria uses a defined set of cellular proteins such as Vasp, actin, Arp2/3, cofilin, capZ, profilin, and alpha-actinin to assemble actin filaments . Any reduction in Listeria tail length or velocity may indicate that LT alters one or more essential proteins necessary for Listeria motility. This finding would then allow us to focus on a defined set of host cell proteins. Cells were infected for 2h with Listeria before addition of anthrax toxi ns for 2h. Velocities and tail lengths were measured as described in expe rimental design by us ing video microscopy. Intracellular Listeria velocities were reduced in anthrax-toxi n treated cells as shown in Fig. 4-6 B (n=66-322, p<0.001). Toxin treatment reduced Listeria velocity by greater than 50% compared to control cells. As shown in Fig. 4-6 A and C, phalloidin staining of Listeria actin tails revealed that toxin treated-cells had s horter tail lengths when compar ed to control cells (control=10.5 m 0.4 m, n=118 versus toxin treated=6.3 m 0.18 m, p<0.0001; n=146). Velocity is directly dependant on the rate of Listeria induced actin assembly, so it wa s no surprise that toxin treated cells had shorter tails and reduced velocities. HeLa cells that were treated with the p38 inhibitor SB203580 also showed reduced intracellular Listeria speeds and shortened actin comet tails. The p38 inhibitor reduced velocities by 50% (n= 128-1189, p<0.0001; Fig. 4-6 B). Tail lengths were reduced from 10.5m 0.4m in control cells to 7.66 0.4m in SB203580 treated cells (Fig. 4-6 C, n=54-63, p<0.0001). It is important to note that the speeds and ta il lengths of the LT-treated
53 cells and the p38 inhibitor treated cells were al most the same, indicating that LT is possibly exerting its effect on Listeria by altering the p38 MAPK pathway. Anthrax LT Blocks Phosphorylation of Hs p27 in HeLa Cells and Human Neutrophils Since I was able to show that LT impairs Listeria intracellular motility in HeLa cells, I performed 2D electrophoresis on HeLa cells treated with buffer alone or with 1 g/ml LT for 12h to look at global protein e xpression. The amount of LT adde d was increased in hopes of enhancing any defects that were present. No significant apoptosis wa s noted to occur when treating HeLa cells overnight with the toxin. As shown in Fig. 47, I identified a polypeptide at 27-kD that was reduced 3.1 times in the presence of the toxin (compare a rrows; representative of 4 independent experiments). The 27-kD protein was identified by MALDI-TOF to be the small heat shock protein Hsp27. To verify the 2D resu lts, I performed wester n blots with both total Hsp27 antibodies and phospho-serine 82 antibodies. Total Hsp27 was the same in both treated and non-toxin treated samples (Fig. 4-8 A). The phospho-serine 82 isoform was completely abolished in anthrax toxin-treate d cells (Fig. 4-8 A). To further verify this result I performed western blots on 2D gels that were transferred to nitrocellulose membranes after toxin treatment. In Fig. 4-8 B, antibodies to total Hsp27 a nd phospho-serine 82 were used to probe the membranes after transfer (Fig. 4-8 B). Based on the phosphorylation status of Hsp27, it would be expected to have different isoe lectric points, and four differe nt phosphorylation states existed when using an antibody for total Hsp27 (Fig.48 B; upper 2 blots). The most phosphorylated form of the protein (most acidic) was decreased by 75% in the t oxin treated samples (Table 4-2; isoform-d) when compared to the toxin-treated samples (Figure 4-8 B; see spots under pH 5.0). In the toxin treated samples the least phosphorylat ed form (most basic) was increased 3.6 times (Table 4-2; isoform-a) when compared to contro l cells (Fig. 4-8 B; spots under pH 6.0). It is
54 important to note that the sum of the intensitie s were almost the same, indicating no significant protein degradation (Table 4-2; Hsp27 total su m). I performed western blots on 2D separated proteins using phospho-serine 82 Hsp27 antibodies. Three different isoforms of phosphorylated protein were identified (Fig. 14B; 2 lower blot s). As shown in Fig. 4-8 B lower panels, P82 Hsp27can barely be detected in LT treated cell s while in control cells P82 Hsp27 was 3 times higher as compared to LT treated cells. 2D ge l western blots were also performed on neutrophil extracts that were treated w ith 500ng/ml of LT for 2h. This concentration of LT was recently shown to have no effect on neutrophil apoptosis . Antibodies directed against total Hsp27 revealed that the most phosphorylated isoform (Fi g. 4-8 E; isform d, and Table 4-2) was reduced in LT treated neutrophils by 3.5 fo ld. Total Hsp27 was not changed (Fig. 4-8 D, and Table 4-2). Data indicates that LT is not directly degr ading Hsp27 (Fig. 4-8 A-E) but is altering the phosphorylation status by disrupting the ce ll-signaling cascade that leads to Hsp27 phosphorylation (Fig. 1-2). Phosphospecific serine-82 antibody did not react in the neutrophil extracts for reasons I do not understand, so phosphorylation was assessed by 2-dimensional electrophoresis only (Fig. 4-8E ). Hsp27 phosphorylation is cont rolled by MAPKAP-2/3 whose phosphorylation is controlled by p38 MAPK (Fig.1-2). To determine if LT was blocking p38 phosphorylation I performed wester n blots on both neutrophil and HeLa extracts that were treated with LT. I found that p38 phosphoryla tion was disrupted by LT (Fig. 4-8 C and F). MAPKK-3/6 phosphorylates p38 MAPK resulting in its activation, but in th e presence of LT p38 MAPK cannot be phosphorylated because of MAPKK-3/6 cleavage (Fig1-2) . Hsp27 Immunoflouresence I also performed immunofluorescence experi ments with antibodies to total Hsp27 and phospho-serine 82 Hsp27 in Listeria -infected HeLa cells and neutr ophils treated with FMLP to examine how these two fo rms of Hsp27 localize to Listeria actin tails and the leading edge of
55 polar neutrophils. An antibody reco gnizing total Hsp27 localizes to the leading edge of polarized neutrophils, where dynamic actin assembly ta kes place (Fig. 4-9 B). An antibody recognizing only phosphorylated Hsp27 on serine-82 localized in the peri-nuclear area , but rarely at the leading edge. (Fig. 4-9 D). This distinct difference in localiz ation led me to believe that phosphorylation of Hsp27 regulates its localization with in the cytoplasm of the cell and that Hsp27 may have distinct func tions depending on phosphorylati on state. Hsp27 localized to Listeria actin comet tails when using an antibody that detects total Hsp27 (Fig. 4-9 F), but when using an antibody that recognizes phosphorylated Hsp27 on serine-82 (Fig. 4-9 H), no such locaization was seen. I hypothesize that non-phosphorylated Hsp27 localizes to Listeria tails and that phosphorylation reverses th e ability of Hsp27 to localize to actin comet tails. I also performed confocal immunofluorescence of neutr ophils using the same two antibodies (Fig. 4-10 A and B). Neutrophils were adhere d to fibronectin-coated plates, before being stimulated with 1M FMLP for 5min. Cells were fixed and perm eabilized before antibody staining. Confocal slices were taken every 0.5m. Two images we re superimposed, one at the adherent surface (green) and another slice 3m hi gher (red). In the cells stained with total Hsp27, green indicates that Hsp27 was seen only at the adherent su rface and indicates an early lamellipod. Yellow indicates that Hsp27 is evenly di stributed in the cell. Red indicates that the adherent surface in these regions contain little Hsp27, and the protein is in the body of the cell. From these cells, it can be clearly seen that nonphosphorylated Hsp27 is localized in the l eading edge of polar neutrophils (Fig. 4-10 A). In Fig. 4-10 B, antibodies that recogni ze P82 Hsp27 were used to stain FMLP stimulated neutrphils. From this image it can be seen that P82 Hsp27 resides mostly in the perinuclear area (centrally locate d yellow) of the neutrophils and does not localize to the leading
56 edge. Most cells are yellow, indicating Hsp27 is cen trally distrubited through the entire Z-axis of the cell. Nonphosphorylated Hsp27 Inhibits Listeria Actin Tail Formation In Rat Brain Extracts Listeria is able to form actin rocket tails in br ain extracts . I have two mutant Hsp27 proteins: one that is pseudo-phos phorylated (Hsp27EE) and one that cannot be phosphorylated (Hsp27AA). In the mutant proteins serines were replaced with glutamic acid (EE) or alanines (AA). Hsp27 is thought to have opposing functions on actin assembly depending on phosphorylation status . I decided to see if Hsp27 AA or EE had any effect on Listeria induced actin tail lengths in rat brain extracts . I chose to use only the AA mutant and not wild type protein because it is possi ble that the wild-type protein could be phosphorylated in the extracts. As shown in Fig. 4-11 A, Hsp27 AA shorte ns Listera rocket tails in a dose dependant response (Bars=SEM of 33-57 tails). Tail length was reduced by greater than 50% when 675g/ml Hsp27 AA was used (p<0.0001;n=46-50 tail s). The EE mutant had an opposite effect on tail length. Addition of 135, 270 and 400g/ml of Hsp27 EE caused a modest increase in tail length (Bars=SEM of 30-34 tails). Data supports earlier findi ngs that indicate Hsp27 inhibits actin assembly when nonphosphorylated . Hsp27 RNAi Blocks LT Activity Hsp27 specific RNAi was used to reduce Hsp27 expression in HeLa cells 48 h prior to Listeria infection. I was able to ach ieve a reduction of greater th an 50% at 48h (Fig. 4-11 B). HeLa cells were also treated w ith LT or buffer alone 1h after Listeria infection. In control cells, LT reduced intracellular Listeria speeds by 50% as previously seen ( p<0.0001, Fig. 4-11 C). HeLa cells treated with Hsp27 RN Ai alone reduced intracellular Listeria velocities by 50%, indicating Hsp27 is involved in Listeria motility (p<0.0001, Fig. 4-11 C). The most surprising result was that when Hsp27 RNAi was used in conjunction with LT no additive effect was seen
57 (Fig.4-11 C). This suggests that LT and the Hs p27 RNAi act on the same target. If Hsp27 was not the target of LT, then I w ould expect an additive effect on Listeria motility, since Hsp27 RNAi reduces Listeria speeds by itself. This led me to hypothesize that Hsp27 is the primary target of LT and that blocking phosphorylation through the p38 MAPK pathway results in an increased amount of nonphosphorylated Hsp27 that w ould be expected to slow the assembly of actin filaments. Actin Kinetics Hsp27 is an Actin Monomer Sequestering Protein Pyrenyl actin was used to examine the eff ects of Hsp27 on purified actin assembly. As actin monomers incorporate into filaments the fl uoresence intensity increases. As shown in Fig. 4-12 A, both wild-type and Hsp27A A inhibit actin assembly when preincubated with G-actin. Hsp27 is sequestering actin monomers, and they are not available for addition to the growing actin filament. Inhibition was concentration de pendant, and total inhibition of assembly was observed at 8-10g/ml (0.04-0.05 M) of Hsp27AA or wild-type Hsp27. The concentration of monomeric actin available for assembly in our experiments was 1.2M (tot al actin concentration of 2M minus 0.8M, the critical concentration of the gelsolin -actin nuclei). Therefore, one homo-octomer of unphosphorylated Hsp27 was capable of sequest ering 24-30 actin monomers. This suggest that each monomer is binding 2-3 ac tin monomers. The filaments were seeded with 1:20 gelsolin, allowing growth from only the pointe d ends of the actin filaments. This procedure eliminated barbed end capping from the experime ntal equation, and the only way polymerization could be slowed was by monomer sequesteration. Previous investigators suggested that Hsp27 is a barbed end capping protein , however I have found Hsp27 to be a potent monomer sequestering protein whose activity is cont rolled by phosphorylation. Hsp27 EE (phosphorylated Hsp27) had no effect on actin assembly in this assay, indicating it is no t a monomer sequestering
58 protein. To test if there was any time depende nce on monomer sequestration, I performed a time course experiment, in which I incubated 13.5 g/ml human Hsp27 with 2M actin monomers before adding to gelsolin-actin nuclei. Addi ng Hsp27 without preincuba ting had no effect on the rate of actin assembly (Fig. 4-12 B). Preincuba ting actin monomers with Hsp27 for 3min gave maximal sequestration, indicati ng slow initial binding between actin monomers and Hsp27. Increasing incubation to 5min did not enhance monomer sequestration. MAPKAP-2 Reverses The Ability of Hsp27 to Sequester Actin Monomers To further prove that phosphorylation cont rols the ability of Hsp27 to sequester monomers, I performed experiments in whic h human Hsp27 was phosphorylated in vitro by MAPKAP-2. As shown in Fig. 4-12 C, MAPKAP-2 could effectively phosphorylate recombinant Hsp27 on serine-82. Forty-five percent of the to tal protein was phosphorylated. To determine if phosphorylation could reverse the ability of Hs p27 to sequester monomers, Hsp27 was incubated with active MAPKAP-2 and ATP for 30 min before adding 1M G-actin for an additional 3 min. Gelsolin:actin seeds were then added to ini tiate actin assembly. As shown in Fig 4-12 D, phosphorylation of Hsp27 completely abolished the ability of Hsp27 to sequester actin monomers (filled grey square; 5.4g/ml and ci rcle; 13.5g/ml). Hsp27 that was not incubated with MAPKAP-2 sequestered actin monomers in a dose dependant response (filled black square; 5.4g/ml and circle; 13.5g/ml). Data data pr ovide conclusive evidence that Hsp27 is a monomer sequestering protein, whose activ ity is controlled by phosphorylation. Non-Phosphorylated Hsp27 Weakly Re duces Barbed End Actin Assembly I next wanted to determine if Hsp27 coul d block barbed end actin assembly, since Bennedorf et al described Hsp27 to be a barbed end capping protein . First, I wanted to determine the Kapp Hsp27 for monomers in the presence of F-actin seeds, in which the barbed ends were free. 54g/ml was required to obtain si gnificant blockage of actin assembly, indicating
59 that Hsp27 cannot sequester monomers very efficiently when the ba rbed ends are free, compared to when the barbed ends are capped (Fig. 4-13 A; filled circles). The Kapp was determined to be 0.25M when the barbed ends were free. When the barbed ends were capped Hsp27 could completely inhibit actin assembly at about 810g/ml (0.04M). This data is in good agreement with Bennedorf et. al, who found th at a 1:1 ratio of Hsp27 to actin was required to block actin assembly when the barbed ends are free . To determine if Hsp27 could cap the barbed ends of actin filaments I performed experiments in which Hsp27 was preincubat ed with spectrin 4.1 or G-actin. Spectrin 4.1 contains small F-actin fila ments that have free barbed ends. If Hsp27 caps barbed ends of the spectrin 4.1 actin filaments, I would expect a decrease in actin nucleation and assembly. As shown in Fig. 4-13 A, spectrin 4.1 nucleation was not blocked by pretreatment with Hsp27, even at high concentrations such as 27g/ml (open squares). This leads us to believe that Hsp27 cannot block actin assembly by capping th e barbed ends of actin filaments. I next looked to see if Hsp27 could increase the critical concentration of monomers in solution when the barbed ends are free. 100g/ml (filled square s) of Hsp27 was found to increase the critical concentration of the monomers from 0.2M to 0.45M (Fig. 4-13 B). 50g/ml (filled circles) only reduced the critical concentration from 0.2M to 0.35M, further indicating a weak rise in the critical. In conclusion, Hsp27 cannot cap the barbed ends of actin filaments but sequesters monomers to inhibit actin assembly. Hsp27 cannot sequester monomers efficiently when the barbed ends of actin filaments are free. Hsp27 AA and EE Slow Disassembly of Actin Filaments I next proceeded to determine if Hsp27 AA or EE had any effect on the disassembly rate of actin filaments. In this assay, actin filaments are sheare d and diluted below the critical concentration to promote disassembly of actin m onomers from the barbed end of the filaments. As shown in Fig. 4-14 A and B, both mutant proteins show a dose dependant slowing of
60 disassembly once the sheared F-actin filaments we re diluted below the critical concentration. The Kapp of the AA mutant was determined to be approximatly 0.32g/ml (1.5nM) while the Kapp of the EE mutant was 0.64g/ml (11nM). I also performed disassembly experiments with human Hsp27 and found it behave identically to Hsp27AA (Kapp= 0.32g/ml, 1.5nM; Fig. 4-14 C). Hsp27 was not found to prevent disassembly of actin monomers from the pointed end of filaments (data not shown).
61 Table 4-1. Percentage of necrotic and apoptot ic neutrophils after LT or buffer treatment. Table 4-2. Densiometry scans of HeLa and Neutrophil Hsp27 after 2D-western blots. Note: Relative integration units Figure 4-1. Western blot show ing MAPKK-1 N-terminal brea kdown when neutrophils were treated with 50ng/ml of LT for 2h. No such breakdown was seen in control cells that were mock treated. This finding indicates th at LT is active and can be expected to exert its effects within 2h in human neutrophils. Condition % Annexin V % Propidium Iodide Control 8.4 4. 9 IgG (500ng/ml) 9.0 5.9 Anti-Fas IgM 19.3 5.5 Lethal toxin (50ng/ml) 8.2 3.7 Lethal toxin+anti-Fas IgM 19.2 4.1 HeLa Hsp27 Isoform-a Isoform-b Isoform-c Isoform-d Sum Control 107 1209 77 490 1883 LT 390 1064 127 134 17 15 Neutrophil Hsp27 Control 1054 244 876 519 2693 LT 1287 364 839 144 263 4
62 Figure 4-2. Anthrax LT blocks neutrophil chem okinesis. A) Neutr ophil velocities after stimulation with 1 M FMLP. Control neutrophils (fil led circles) exhibited higher velocities when compared to toxin tr eated cells (open circles LT 50ng/ml 2h). Velocites were recorded every 20 sec, as described in experimental design. (SEM 1114 neutrophils). B) Neutrophil veloc ities as a function of time after 1 M FMLP stimulation. Control neutrophils (black bars ) exhibited significantly faster velocities than toxin treated neutrophils (grey bars LT 50ng/ml 2h). This decrease in velocity lasted throughout the entire 12min obse rvation period (left insert; SEM 135-210 measurements, n=14 cells/period; p< 0.0001) . Neutrophils treated with 50ng/ml LT for 1h exhibited a significant decrease in velocities (SEM 230-240 cells; n=30 cells; p<0.001) also, but the reduction in velocity was only 33% as compared to neutrophils that were treated with toxin for 2h, which e xhibited a 66% decrease in velocity (right insert). C) Phase contrast micrograph of neutrophils after 5min stimulation with 1 M FMLP. Arrows indicate the polarity of th e cells. D) Phase contrast micrograph of toxin treated neutrophils. Note the lack of polarized morphology. Only one cell in the micrograph exhibited a polarized mor phology. E) Quantification of neutrophil polarity in mock (black bars), or LF+P A 50ng/ml treated neutrophils (grey bars 2h treatment). Control neutrophils formed signi ficantly more polarized structures when compared to toxin treated cells (p<0.0001, Fischerâ€™s exact test; left insert). No difference in polarity was seen when neut rophils were treated for 1h (right panel; p=0.66 Fischerâ€™s exact test). 37-112 cells were analy zed in each condition.
63 Figure 4-3. Anthrax LT blocks neutrophil chem otaxis. A) Neutrophil chemotaxis using a microinjection needle containing 10 M FMLP. Control cells formed polar structures while crawling toward the needle. Cells tr eated with 50ng/ml of LT for 2h, failed to polarize and crawl toward the needle. Arrows denote lamellipod forming from the control cell. B) Quantificati on of neutrophil velocity. Cont rol neutrophils (black bar) had significantly higher velocities when comp ared to cells that were treated with 50ng/ml LT for 2h (grey bars; SEM 131 for c ontrol and 105 for toxin treated cells; 30 cells, 5 experiments).
64 Figure 4-4. Anthrax LT blocks neutrophil actin assembly. A) Relative F-actin content as determined by FACS analysis of phalloid in stained neutrophils (5,000-10,000 cells analyzed, SEM=3 experiments). Toxin-treate d neutrophils consiste ntly exhibited a reduction in F-actin when compared to mo ck treated cells (filled circles). B) Neutrophils were treated with 5, 50,or 500ng/ ml LT (grey bars) for 2h, as described in experimental methods. At 30s toxin tr eated neutrophils exhi bited a 50% reduction in F-actin when compared to control neutrophils. C) Neutrophils treated for 1h with 50ng/ml LT (open circles) did not exhibit any defect in F-actin assembly when compared to control neutr ophils (filled circles SEM=3 experiments). D) Relative Factin content as determined by the triton-insoluble cytoskeleton assay. Densitometry scanning of the resulting 43-kD protein was performed to determine F-actin content. Toxin treated neutrophils (open circles) exhi bited a defect in actin assembly at both 20 and 40s when compared to control cells (fi lled circles mean of 2 experiments).
65 Figure 4-5. p38 MAPK is required for neutrophil actin assembly. A) Neutrophil actin assembly in the presence of the Erk inhibitor PD 98059. Neutrophils were pretreated with 100 M PD98059 for 10min before stimulation with 1 M FMLP for various time points. Cells treated with the Erk inhibitor (open circles) did not exhibit a defect in actin assembly when compared to control neutrophils (mean of 2 different experiments). B) Neutrophil actin assembly after treatment with the p38MAPK inhibitor SB203580, followed by stimulation wi th 1M FMLP. Cells were fixed and stained with Alexa-488 phalloidin, followed by flow cytometric analysis. Neutrophils were treated with buffer alone or SB203580 (100M) for 10min before stimulation. SB203580 reduces actin assembly in a simila r manner to LT (open circles). Mean of two separate experiments. A B
66 Figure 4-6. Anthrax LT blocks intracellular Listeria actin assembly. A) HeLa cells were infected with L. monocytogenes for 2h before the addition of anthrax lethal toxin (50ng/ml). Phalloidin staining was carried out as desc ribed in the materials and methods. Left images show actin comet tails of Listeria in control cells, while the right images show LT treated cells. Notice relatively long actin tails when untreated, as compared to LT treated cells. B) Quantification of intracellular Listeria velocities in control cells (open bar), LT-treated cells (black bar), and SB203580 treated cells (grey bar). LT and SB203580 reduced intracellular Listeria velocities by more than 50%. Reduction in speed was extremely significant (p< 0.0001). C) Quantification of intracellular Listeria actin tail lengths. Cont rol cells (open bar) had significantly longer actin rocket tails (10.5 m 0.4 m, n=118) versus toxin (bl ack bar) treated cells (6.3 m 0.2 m, p<0.0001; n=146) and SB203580 treated cells (7.66 0.4m). This result is consistent with a reduction in intracellular speeds as shown above. Tail lengths were measured after phalloidin staining by us ing Metamorph 4.0 as described in the methods section.
67 Figure 4-7. 2D-SDS PAGE of control HeLa cells and LT (1g/ml) treated cells . A total of 400 g of protein was loaded onto a 3-11 IP G strip. Strips were rehydrated for 12h and focused for 6h as described in experi mental design. Polyacrylamide gels were used for 2nd dimension separation. Gels were run at 150V for 6h before silver staining. Arrow indicates the position of Hsp27 spot that was identified by MALDITOF. Control cells had 3.1 times more of the phosphorylated Hsp27 isoform. Gels representative of 4 independent experiments.
68 Figure 4-8. LT blocks Hsp27 phosphorylation. A) We stern blot analysis of control and LT (1g/ml) treated HeLa cells. Total Hsp27 was the same, while serine 82 phosphorylation was completely abolis hed in the toxin treated cells. B) Western blot analysis of control and LT tr eated HeLa cells after 2D separation as described in methods. Anti-Hsp27 antibodies ( upper 2 panels) revealed th e presence of 4 different Hsp27 isoforms that differ by pI, which is a function of phosphorylation status. Arrow shows most phosphorylated (acidic) fo rm which is abundant in control cells, but nearly absent in toxin treated ce lls. Anti-phospho serine-82 Hsp27 antibodies (lower 2 panels) revealed the presence of 3 different Hsp27 isoforms that are phosphorylated at serine-82. Control cel ls have an increased proportion of phosphorylated Hsp27, when compar ed to toxin treated cells. C) p38 western blot analysis of control and LT (1g/ml) tr eated HeLa cells. Total p38 was the same, while phosphorylation was completely abolis hed in the toxin treated cells. D) Western blot analysis of ne utrophil extracts that were treated with buffer or LT (500ng/ml for 2h). Total Hsp27 was the same in both the control and treated samples. E) Western blot analysis of control and LT treated neutrophils after 2D separation. Anti-Hsp27 antibodies revealed the presence of 4 different Hsp27 isoforms. Arrow shows most phosphorylated (acidic) form, which is abundant in control cells, but nearly absent in toxin treated cells. F) p38 western blot analysis of control and LT (500ng/ml; 2h) treated neutrophils. Phosphor ylation of p38 was found to be reduced in LT treated samles.
69 Figure 4-9. Hsp27 localizes to Listeria actin tails and the leading e dge of neutrophils. A) Phasecontrast image of FMLP stimulated neut rophils. Neutrophil movement is in the direction of the arrows. B) Fluorescent micrograph of anti-Hsp27 stained neutrophils. Hsp27 localized to the lamellipodia (leadi ng edge) of polarized neutrophils. Neutrophils were fixed and permeabilized be fore Hsp27 staining as described in the methods. C) Phase-contrast image of FMLP stimulated neutrophils. Neutrophil movement is in the direction of the arro ws. D) Fluorescent mi crograph of anti-Hsp27 serine-82 stained neutrophils. Phosphorylated Hsp27 localized to the perinuclear area of polarized neutrophils. Note, that phosphorylated Hsp27 localization is very different from that of non-phosphorylated Hs p27. E) Phase-contrast image of HeLa cells that were infected with Listeria for 3-4h prior to fixatio n. Arrows indicate phase dense actin tails of motile Listeria . F) Fluorescent micrograph of Hsp27 (total) stained HeLa cells that were infected by Listeria . Cells were infected 3-4h before fixation. Arrows show localization of Hsp27 to the Listeria actin comet tails. G) Phase-contrast image of HeLa ce lls that were infected with Listeria for 3-4h prior to fixation. Arrows indicate phase dense actin tails of motile Listeria . H) Fluorescent micrograph of phospho-serine 82 Hsp27 staine d HeLa cells that were infected by Listeria . Cells were infected 3-4h before fixa tion. Arrows show no localization of phopho-serine 82 Hsp27 to the Listeria actin comet tails.
70 Figure 4-10. Confocal immunofluor esence of polarized neutrophils. A) Neutrophils were adhered to fibronectin-coated plates and stimulated with 1m FMLP for 5min before being fixed with 3.7% formaldehyde. Cells we re permeabilized with 0.2% triton X-100 before being stained with total Hsp27. Conf ocal images were collected every 0.5m. Two cross-sections were superimposed in th is figure. Green represents the adherent surface of the cells, whereas re d indicates the apical surfac e of the cell 3m above the green cross section. Note that Hsp27 is f ound in the lamellipod of polar neutrophils. B) Confocal images of ne utrophils that were staine d with P82 Hsp27 antibody. P82 Hsp27 was centrally located and distributed evenly throughout the Zaxis of the cells (yellow color). P82 Hsp27 was not found at the leading edges of polar neutrophils. For color reference see above description. A B
71 Figure 4-11.Hsp27 siRNA blocks the action of LT . A) Various concentrations of Hsp27 AA and EE were added to rat brain extract that was supplemented with Listeria Bug 666 and rhodamine labeled actin. 2l of extract wa s then sandwiched between 1% BSA coated slides and coverslips for 30min. Tails were observed using fluorescent microscopy, and tails were measured using Metamorph 4.0. Hsp27 AA decreased Listeria tail lengths in a dose dependant manner (SEM of 33-57 tails). 675g/ml reduced Listeria tail lengths by greater than 50% (p< 0.0001; n=46-50 tails). Hsp27 EE modestly increased Listeria tail lengths (SEM of 30-34 tails). All c oncentrations used increased tail lengths in a similar fashion. B) We stern blot analysis of Hsp27 after RNAi treatment using both control and Hsp27 speci fic anti-sense RNA (upper two panels). Lower figure shows relative concentration of Hsp27 after 48h of RNAi treatment. A reduction of greater than 50% was found when Hsp27 specific anti-sense RNA was used instead of scrambled control anti-sense RNA. C) HeLa cells were treated with control siRNA or Hsp2 7 RNAi, 48h before Listeria infection. Cells were infected with Listeria for 1h before the addition of buffer or LT (50ng/ml) for 2 more hours. Intracellular Listeria velocities were analyzed by using Metamorph 4.0. Hsp27 and LT reduced Listeria velocities by 50% when added independently. The addition of LT and Hsp27 siRNA together did not produce an additive effect, thus leading us to conclude that Hsp27 is the target of LT. B C
72 Figure 4-12. Hsp27 sequesters actin monomers. A) 120nM of gelso lin:actin (1:20) seeds were used to initiate actin assembly of 2 M G-actin that was preincubated with Hsp27 mutant and wild-type proteins for 3mi n. Rates were calculated from the linear portions of the assembly curves. Both wild-type Hsp27 and Hsp27AA reduce pointed end actin assembly in a dose dependant manner. The Kapp was determined to be 0.04M if non-phosphorylated Hsp27 is consid ered an octamer. Hsp27EE had little effect on actin assembly. Since the barbed ends are capped, the reduction in actin assembly can only be a result of monomer sequestration. B) 2M G-actin was preincubated with 13.5g/ml human Hsp27 before being added to 120nM gelsolinactin (1:20) filaments to in itiate actin assembly. Hsp27 sequesters actin monomers in a time dependant manner as shown by a decrease in assembly rates. Maximum sequestration occurred with 3min in cubation of monomers with Hsp27. C) MAPKAP-2 can phosphorylate recombin ant Hsp27 on serine-82. Lane without MAPKAP-2 shows no phosphorylation of Hs p27, but when MAPKAP-2 is added phosphorylation on serine-82 occurs. 0.1g of MAPKAP-2 was used to phosphorylate 0.67g of Hsp27 for 30min. D) Phosphorylation of Hsp27 reverses its ability to sequester monomers. Hsp27 that is phosphorylated by MAPKAP-2 can no longer sequester monomers (5.4g/ml grey square; 13.5g/ml grey circle). 1g of MAPKAP-2 was incubated with either 5.4g/ml or 13.5g/ml of Hsp27 prior to addition of G-actin. Hsp27 that was not phos phorylated could sequester monomers in a dose dependant response (5.4g/ml f illed square; 13.5g/ml filled circle).
73 + MAPKAP-2 MAPKAP-2 + MAPKAP-2 MAPKAP-2 Anti-Hsp27 Anti-Ser-82 Hsp27 0.0 100.0 200.0 300.0 400.0Rate units/s 0 50 100 150 200 250 300 350Relative Fluorescence060120180240300Time,s C A D No 0 min 1min 3 min 5min Hsp27 B
74 Figure 4-13. Hsp27 weakly sequesters actin monom ers when barbed ends are free. A) 1M G actin was added to 120nM spectri n-actin filaments to initiate actin assembly in the presence of various concentrations of human Hsp27. Human Hsp27 at concentrations of 13.5g/ml, 27g/ml, and 54g/ml weakly blocked barbed end growth (filled circles). The Kapp was determined to be 0.25 M, which was found to be much weaker than when the barbed ends are capped (0.04M). Hsp27 cannot cap the barbed ends of actin filaments. Hsp27 was pre-incubated with spectrin 4.1 or G-actin for 3min prior to initiati on of polymerization. Pre-incu bation of spectrin 4.1 with 6.75g/ml, 13.5, and 27g/ml (open squares) showed no reduction in actin assembly. B) 2M F-actin was polymerized overnigh t. Different amounts of G-actin were incubated with human Hsp27 before the addi tion of 120nM F-actin seeds to initiate polymerization. 100g/ml Hsp27 (filled squares) increased the critical concentration of actin monomers in solution from 0.2M to 0.4M. 50g/ml of Hsp27 was found to increase the critical conc entration from 0.2M to 0.35 M (filled circles), when compared to control samples (open circles). 0 100 200 300 400 500 600 700 800 900 1000 0102030405060708090100Concentration, g/ml B A Assembl y Rate ( units/s )
75 Figure 4-14. Hsp27 slows the disassembly of actin filaments. A) Hsp27 AA used at concentrations of 0.320g/ml (filled circ les), 0.160g/ml (filled squares), and 0.080g/ml (filled triangles). All concentr ations of Hsp27 slowed the rate of fluorescence decrease, when compared to c ontrol sheared filaments (open circles) The Kapp was determined to be 0.32g/ml (1.5nM). B) Hsp27 EE was found to slow the disassembly rates of actin filaments diluted below the critical concentration at Hsp27 concentrations of 0.640g/ml (filled circles), 0.320g/ml (filled squares) and 0.160g/ml (filled triangles), when compared to control filaments (open circles). The Kapp was determined to be 0.64g/ml (11nM). C) 2M spectrin-actin was mechanically sheared and diluted below th e critical concentra tion (100nM), in the presence of 0.320g/ml (filled squares), 0.160g/ml (filled triangles) and 0.080g/ml (filled crosses) of human Hsp27. Human Hs p27 slows disassembly rates of actin filaments in a dose dependant manner as shown by retention of fluorescence, when compared to control filaments (open circles). The Kapp was determined be the same as for Hsp27AA (0.32g/ml; 1.5nM). 0 250 500 750 1000Relative Fluorescence060120180240300Time, s A B C
76 CHAPTER 5 DISCUSSION LT Paralyzes Neutrophil Actin Assembly I showed that 50ng/ml of LT does not induce any significant apoptosis or necrosis in neutrophils that were incubate d with the toxin components for 2h. This was an important finding because LT causes macrophage apoptosis . R ecently, it has been shown that neutrophils do not undergo LT induced apoptosis, even when high concentrations of LT are used for long periods of time . LT at low concentrations (50ng/ml) efficiently clea ves the N-terminus of MAPKK-1 in human neutrophils. Earlier neutroph il studies examining anthrax toxin components did not have the capability to monitor MAPKK1 cleavage because the target of LT was not known . MAPKK-1 cleavage is usef ul as a gauge for toxin activity. For neutrophils to properly kill invading pa thogens, they must first undergo chemotaxis to sites of infection. Any bloc k in neutrophil chemotaxis woul d severely impair the innate immune system. Neutrophils are the first responder s to sites of infection, thus a lack of an efficient neutrophil response severely compromi ses the hostâ€™s chance of survival in severe infections, such as anthrax. Neutrophils must also undergo chemotaxis from the bone marrow to the blood stream during development. Any paralysi s of neutrophil chemotax is would also severly disrupt this process. I showed that anthrax LT paralyzes neutr ophil chemokinesis and chemotaxis in response to FMLP. Neutrophils that were trea ted with LT also failed to polarize when exposed to FMLP. These findings contra dict previous work that was performed nearly 20 years ago. Wade et al, showed that LT increased the speed s of human neutrophils that were exposed to chemoattractant gradients. I believe there are pot entially two reasons for the differences between our findings and Wade et al. Firs t, in the previous study LT was incubated with the neutrophils
77 for only 1h, instead of 2h as in our study. I have f ound that the full effect of LT takes at least 2h. Incubating neutrophils with LT for only 1h failed to cause a defect in ne utrophil polarization and resulted in only a small defect in chemokinesis. These effects were very different than those observed when LT was applied for 2h. Our findings indicate that paralysis of neutrophils by LT requires at least 2h. Secondly, the target of LT was not known, so there was no way to determine the activity of the toxin. Also, th e previous researchers had no way to measure the kinetics of MAPKK cleavage in neutrophils to know if or when the toxin was active in the cytoplasm. It is also of interest to note that LT blocked directional motility (chemotaxis) much more effectively than random motility (chemokinesis). Chemotaxis was almost completely abolished, while chemokinesis was reduced by only 50%. This indi cates that LT may be able to impair chemotaxis by blocking the ability of neutrophils to sense chemoattracta nt gradients. During chemotaxis certain proteins such as the sma ll GTPases Rac and Cdc42 localize to the leading edge of neutrohils (lamelliapod), and the small GTPase Rho localizes to the back or uropod of the cell [143,144]. Rac and Cdc42 ar e thought to aid in actin asse mbly and attachment to the substratum, while Rho is thought to aid in cont raction and detachment in to the substratum [143,144]. It is possible that LT coul d block the localization or activ ation of polar proteins such as those mentioned above, and could result in random motility instead of directional motility. For cell motility to occur, actin monomers must be assembled into filaments to provide the force necessary for movement. Since LT im paired neutrophil motility, I hypothesized that actin assembly would be reduced. I showed by phallo idin staining that LT impairs the ability of neutrophils to assemble actin in response to FM LP. This is the first time that anyone has shown LT to have an effect on the actin cytoskelet on. Not only was a reduction noted, but there was also a delay in the kinetics of actin assembly, w ith an actual decrease shown at 5s in the toxin
78 treated samples. I used concentrations of LT from 5-500 ng/ml, with no appreciable difference being noted between the lowest a nd highest concentrations of t oxin. After 1h of treatment with LT, no effect on F-actin assembly was noted. Th is correlates well with our chemokinesis and polarization data, indicating that 2h may be required to achieve a reduction in actin assembly. I verified these results by performing the triton insoluble cytoskeleton assay in which actin filaments are pelleted by centr ifugation after FMLP treatment. This assay requires longer actin filaments to pellet, whereas phalloidin stains al l actin filaments, no matter the size. Therefore, this is thought to be a stricter invivo test for the measurement of actin filament assembly. I found that LT markedly reduced F-actin assemb ly as measured by this assay, supporting the validity of our phalloidin results. One interesting aspect of ou r data was that I could never completely block actin assembly with LT treatme nt. This would indicate that neutrophils have multiple signaling pathways that lead to actin assembly, and LT may only be affecting one of these pathways. To determine if MAPKK inactiv ation could cause a reduction in actin assembly, I treated neutrophils with the Erk inhibitor PD98059 and the p38 inhibitor SB203580. I determined that the Erk inhibitor PD98059 had no effect on FM LP induced actin assembly, whereas the p38 inhibitor SB203580 inhibited neutroph il actin assembly in a manner th at was nearly identical to that of LT. This leads us to conclude that LT may block actin assembly by inhibiting the p38 MAPK pathway by cleaving MAPKK-3/6. The p38 MA PK pathway has been shown to be very important for proper neutrophil function  . Granluocyte-macrophage colony stimulating factor, LPS, and PAF (platelet activating factor) activate the p38 MAPK pathway, indicating a role for p38 signaling during the inflammato ry response [145,146]. Recently, p38 MAPK has
79 also been shown to be involved in the neutrophil oxidative burst that is requi red to kill invading pathogens . In conclusion, I have found that anthrax LT is able to paralyze neutrophil chemotaxis and actin assembly. These findings suggest that LT may be able to paralyze the innate immune system by inducing macrophage apoptosis and by inhibiting neutrophil chemotaxis. This may also help explain why in dividuals with inhalation anthrax de velop septic shock. Patients exhibit extreme bacteremia shortly after anthrax infection, indicating th at innate immunity is profoundly impaired . Adaptaive immune responses usua lly require 10-12 days be fore an effective IgG response can begin to control the infection . For most patient s this is simply too long, and they expire before an adaptive response even occurs. Anthrax Lethal Toxin Blocks L. monocytogenes Intracellular Motility I showed that LT and SB203580 have the ability to bl ock intracellular Listeria motility. Listeria is a useful model for studying actin assemb ly, because it uses the hostâ€™s cellular actin machinery to propel itself trough the cytoplasm. Listeria also bypasses the complicated outside in signaling pathways that most cells use to initiate actin assembly. Therefore, Listeria allows us to simplify our experiments by reducing the number of potential proteins anthrax might target. Listeria uses a defined set of cellu lar proteins such as VASP, CapZ, profilin, and the Arp2/3 complex to initiate actin assembly . One problem with this approach is that no one has ever shown p38 MAPK to be involved in Listeria motility. However, based on our inhibitor data, I believe this pathway is likely to be involved in the re gulation of Listera actin assembly. Since I showed LT and SB203580 could slow Listeria actin assembly in HeLa cells, this led us to use HeLa cells as a model to look for specific cha nges in protein expressi on or post-translational modification by using 2D-analysis, paying particular attention to spots at the pI and molecular weight of proteins known to be required for Listeria motility. While performing 2D-analysis I
80 found one particular polypeptide of 27kD that was absent upon treatment with LT. To our surprise this protein did not match the pI or molecular weight of any of the Listeria specific proteins. MALDI-TOF identified this polypeptide to be the small heat shock protein27 (Hsp27), whose activity is cont rolled by p38 MAPK [148-152]. Hsp27 localized to Listeria actin tails when using an antibody that recognizes total Hsp27. When using a phospho serine-82 antibody, no su ch localization was seen. This led me to believe that only nonphosphorylat ed Hsp27 can localize to Listeria actin rocket tails. This may indicate that Listeria can control the amount of Hsp27 present in the actin tails by inducing phosphorylation of Hsp27 through a p38 MAPK dependant manner, since SB203850 slows Listeria speeds. LT and SB203580 disrupt Listeria actin assembly by blocking p38 MAPK activity, which would be expected to resu lt in increased nonphosphorylated Hsp27. This indicates that Listeria needs a balance of nonphophorylated and phosphorylated Hsp27, and too much or too little of either may antagonize actin assembly. Listeria does seem to require Hsp27 in some capacity because Hsp27 specific siRNA reduced intracellular velocities by about 50%, which was the same amount of reduction achieved by LT and SB203850 treatment. This supports a role for Hsp27 in intracellular motility, a nd our current data led us to believe that too much non-phosphorylated Hsp27 is harmful for Listeria actin motility. Both LT and SB203580 increase the amount of nonphophorylated Hsp27 present by blocking p38 MAPK activity. I showed that Hsp27 was the target of LT by usi ng Hsp27 specific siRNA. No additive effect was noted when using both LT and Hsp27 siRNA, indi cating that the siRNA and LT act on the same pathway. This is the first time that anyone has shown Hsp27 to be in Listeria actin tails. The role of Hsp27 in Listeria motility will be discussed later.
81 Hsp27 Function Anthrax LT had no effect on the amount of Hsp27 present in both neutrophils and HeLa cells, but did block the phosphorylation of Hsp27. I conclude this is mediated through LT cleavage of MAPKK-3/6, which controls p38 MAPK activity. I showed that p38 MAPK was not properly phosphorylated in LT tr eated neutrophils and HeLa cel ls, thus further implicating MAPKK-3/6 cleavage as the culpript. When perfo rming 2D-gels I found 4 di fferent isoforms of Hsp27 that exists (A, B, C, and D). This corres ponds well with previous work that has shown 4 different isoforms of Hsp27 to exist in HeLa cells that were treated with arsenite, TNF, and PMA . Three of these isoforms were phosphorylated (B, C, and D) . Hsp27 is phosphorylated on serines 15, 78, and 82 . This phosphorylation is mediated by the serine/threonine kinase MAPK AP-2, whose activation is de pendant on p38 MAPK . In the cell, nonphosphorylated Hsp27 exist as large oligomers upwar ds in size to 700-800 kD . Oligomeric contacts are mediated by the -crystallin domain that resides in the carboxy terminus of Hsp27 . Control of oli gomer size is mediated by phosphorylation of Hsp27. Hamster Hsp27 has only two phosphorylation sites, serines 15 and 90, instead of the three in human Hsp27, thus manipulating ha mster Hsp27 is somewhat easier. . Phosphorylation of Hsp27 by MAPKAP-2 breaks up the large oligometric structures into dimers . Phosphorylation of hamster Hsp27 on serine 90 is sufficient for dissociation of the large 700 Kd structures into dimers [ 154]. Phosphorylation of Hsp27 on serines 15 and 90 is required for full thermoprotection . Thermoprotection of cells by Hsp27 requires phosphorylation which results in changes of the oligomeric stru cture to smaller Hsp27 dimers . Chinese hamster cell lines that were transfected with a pseudophosphorylated mutant Hsp27 protein showed increased resistance to heat shock and microfilament stabilization . In contrast,
82 cells that were transfected with a non-phosphor ylatable mutant Hsp27 could not survive heat shock and there was a decrease in microfilament organization . Hsp27 can regulate the actin cytoskeleton [142,150]. Hsp27 has been shown to block barbed end actin assembly only when non-phosphoryl ated . It was concluded from data that Hsp27 was an actin barbed end capping protein . The physiologic importance of this finding must be called into question, because a 1: 1 ratio of Hsp27 to actin was required to get a significant blockade. Cytochalasin D treatment of smooth muscle cells displaced Hsp27 from the leading edge of the cells, allo wing the authors to postulate that Hsp27 was a capping protein . Antibody staining reveal ed that only non-phosphorylated Hsp27 localized to the leading edge of smooth muscle cells, while phosphor ylated Hsp27 localized to the base of the lamellipodia . Also of interest, is the fact that phospho-p38 MAPK tr ansiently localizes to the front of lamellipodia, while p38 MAPK remain ed phosphorylated at the base of lamellipodia . The authors concluded th at it was p38 MAPK that was controlling Hsp27 phosphorylation through MAPKAP-2 phosphorylation at the leading edge of smooth muscle cells . Chaperone activity requires both serine 15 and 90 phosphorylation . Hsp27 is thought to interact with denatu red proteins through the WD/EPF domain, which is thought to tightly interact with the -crystallin domain . Intramol ecular interactinons between the WD/EPF domain and the -crystallin domain is also though t to be disrupted by serine 90 phophorylation, releasing the WD/EPF domain and a llowing it to interact with any target proteins once serine 15 is phophorylated . Se rine 15 phosphorylation is required for full chaperone activity.
83 Hsp27 and Neutrophils Relatively little work exists on the role of Hsp27 in neutrophils . It is of interest to note that FMLP exposure activates MAPKAP-2, lead ing to the phosphoryla tion of Hsp27 . A one-minute exposure of neutrophils to FMLP resulted in maximum MAPKAP-2 phosphorylation . In our work with neutrophils and LT it is important to note that FMLP was used as the chemoattractant, providing further evidence that Hsp27 function is important for FMLP mediated chemotaxis, specifically directiona l motility. Further evidence for this can be found in our data showing that Hsp27 localizes to the lamellipodi a in polarized neutr ophils. Phosphorylation on serine-82 reversed this locali zation, leading us to believe th at non-phosphorylated Hsp27 may be important for sensing chemoattractant gradients. More recent work has implicated a potential role for Akt (protein kinase B) phopshorylati on of Hsp27 in human neutrophils . Akt is a serine/threonine kinase whose activ ity is controlled by PI-3K . The major role of Akt in the cell is to promote survival and differention in re sponse to external stimul i . Akt is able to phosphorylate Hsp27 only on serine 82 and not 15 and 78 . Furthermore Hsp27, p38, MAPKAP-2 and Akt exists as a complex in th e cytoplasm . The interaction between Hsp27 and Akt is required for Akt activation and subseque nt cell survival . It was also shown that sequestration of Hsp27 away from Akt resulted in enhanced ne utrophil apoptosis and that Akt phosphorylation of Hsp27 on serine 82 caused Hsp27 to dissociate from the complex mentioned above. Data indicates that Hsp27 may someho w regulate neutrophil apoptosis through an unknown manner. Our data indicate that LT does not cause significant apoptosis or necrosis in human neutrophils even though p38 MAPK mediated Hsp27 phosphoryl ation is blocked. To date LT has not been shown to have any effect on Ak t phosphorylation. It is important to note that neutrophils are resistant to LT induced apoptosis whereas, macr ophages are extremely sensitive
84 to toxin exposure [40,139]. The reasons for this are not fully understood, and more work is needed to determine the exact mechanism. Kinetic Analysis of Hsp27 with Actin I demonstrated that Hsp27AA could reduce actin assembly using Listeria and rat brain extracts. In this assay, large qua ntities of Hsp27 had to be used to get an appreciable effect on Listeria tail formation. This result may be because Hsp27 was being regulated in a negative sense by another protein in the rat brain extracts. I know that phosphorylation was not likely because I used Hsp27 AA, which has serine 15 a nd 90 both removed and replaced with alanines. Hsp27 EE had no effect on Listeria tail length, in fact Hsp27 EE slightly enhanced tail formation. The inhibition of actin tails by Listeria when using non-phosphorylated Hsp27 also corresponds well with th at of Benndorf et al, who also showed that only non-phophorylated Hsp27 could block actin assembly . I showed that Hsp27 could sequester monomers in a dose dependant manner only when it was nonphosphorylated. Phosphorylation revesed the ability of Hsp27 to sequester monomers. I found that one homo-octomer of Hsp27 could sequester 24-30 actin monomers, indica ting each Hsp27 bound 3 actin monomers. These monomer sequestering assays were performed using gelsolin seeded actin filaments, so that the barbed end could not be accessible to Hsp27. This would eliminate barbed end capping from the experimental equation, leaving only pointed end capping and mono mer sequestering as means in which Hsp27 could block assembly. I found no evid ence that Hsp27 could cap the pointed ends of actin filaments (data not shown), meaning that Hsp27 is a monomer sequestering protein. Benndorf et al also preincubated G-actin with Hsp27 for 5min before the addition of actin nuclei. This means that Hsp27 was probably sequestering actin monomers before the seeds were added. Thus, it is likely that the results obtained by Benndorf et al, were due to monomer sequestering and not due to capping, indicating the data was probably misinter preted. In fact, I found that
85 Hsp27 had no effect on actin assembly when pr eincubated with spectrin 4.1, which supplies barbed ends to nucleate assembly. Nonphos phorylated Hsp27 increased the critical concentration of actin monomers in solution. This indicates that Hsp27 was weakly removing available monomers from the solution, resu lting in higher criti cal concentrations. I developed a model that depicts how I be lieve Hsp27 monomer seque stering activity is controlled (Fig 5-1). Hsp27, when it is not phosp horylated, has the ability to sequester actin monmers, but once it becomes phosphorylated by MAPKAP-2 the monomers are released and can be added to the barbed ends of actin fila ments. A phosphatase then removes the phosphates from Hsp27 allowing it to sequest er monomers and repeat the cycle again. This model would help explain the presence of Hsp27 in the actin tails of Listeria , Hsp27 may be delivering monomers to the actin tails of Listeria , and an unknown bacterial or cellular kinase could potentially phosphorylate Hsp27. The fact that LT and SB2035850 slow Listeria motility, suggests a cellular kinase such as MAPKAP-2 is involved, but no data exist on whether or not MAPKAP-2 is recruited to Listeria actin tails. This may also explain why I see no phosphorylated Hsp27 in the Listeria actin tails, because once Hsp27 is phosphorylated, it releases the monomers and is no longer neede d. Once this occurs the phosphates are removed by a phophatase, and Hsp27 can sequester new monomers for delivery. LT blocks the phosphorylation of Hsp27 by cleaving MAPKK-3/ 6, resulting in exce ss non-phosphorylated Hsp27 and SB2035850 increases the amount of non-phosphorylated Hsp27 by blocking p38 MAPK activity. Both of these compounds slowed Listeria motility, essentially by the same mechanism. Without the ability to phosphorylat e Hsp27, the monomers cannot be released to incorporate in the Listeria tails. The same process may be occurring in neutrophils that have been stimulated by FMLP. In this case, non-phos phorylated Hsp27 localizes to the leading edge
86 of the cell and becomes phosphorylated by MAP KAP-2, causing the release of monomers and allowing them to be incorporated on the barbed ends of the actin filaments. After this, phosphorylated Hsp27 then leaves the leadi ng edge and becomes dephosphorylated by a phosphatase, so that it can sequester more actin monomers. Hsp27 would then return to the leading edge with actin mono mers in hand. FMLP can activat e MAPKAP-2, resulting in Hsp27 phosphorylation . This would favor the rel ease of actin monomers by Hsp27. FMLP also stimulates actin nucleation by the Arp2/3 co mplex and barbed end uncapping by PI(4,5)P2 (Fig. 1-3). This means that three different pathways which promote actin assembly are activated by FMLP allow dynamic actin polymerization in re sponse to FMLP. Thus, FMLP receptor binding causes nucleation, monomer release, and uncappi ng of preexisting filaments: all of which promote actin assembly and neutrophil motility. LT could therefore block neutrophil motility by producing an excess of non-phosphorylated Hsp27 th at cannot release m onomers for new actin assembly. How non-phosphorylated Hsp27 is targeted to the leading edge of neutrophils and the actin tails of Listeria remains a mystery. Hsp27 also slowed the disassembly of actin filaments that were sheared and diluted below the critical concentration. This activity was not dependant on phosphorylation status. Once the filaments are sheared and diluted, monomers will rapidly dissociate from the free barbed ends of the filaments. Hsp27 could slow the f ilament disassembly by either barbed end capping or by binding to the sides of the filaments a nd providing stabilization so that the monomers cannot exit the filament. I alrea dy know that Hsp27 is not capping the barbed ends of actin filaments, so my findings are most consistant with the latte r option. Phosphorylated Hsp27 is thought to be thermoprotective by binding to the sides of actin filaments and stabilizing them during times of stress, which fits well with our disassembly data . In our assays,
87 nonphosphorylated Hsp27 bound to filaments with a higher affinity than the phosphorylated isoform, but cellular data suggest non-phosphoryl ated Hsp27 not to be th ermoprotective . It is possible that under conditions present in our assays that non-phosphorylated Hsp27 interacts with actin filaments in a way that is similar to the phosphorylated form, however the unphosphorylated form would be expected to main tain a higher concen tration of monomeric actin by sequestration. Both Hsp27 AA and EE slow the off-rate of actin monomers. Slowing the off rate of monomers would be expected to decr ease the critical concentration (Cc=off rate/on rate) of monomers in solution. This would he lp to promote actin assembly, and may be responsible for weak mono mer sequestering when the barbed ends are free. I also noticed a difference in the ability of Hsp27 to sequester monomers depending upon which end of the actin filament is free. It ap pears from our data that Hsp27 can sequester monomers much more effectively when the barb ed end of actin filaments are capped (Compare Fig. 4-12 A and 4-13 A). When the ba rbed ends are capped, 8g/ml (Kapp=0.04M) Hsp27 can block virtually all actin assembly (Fig. 4-12 A), whereas when the barbed ends are free, 54g/ml (Fig. 4-13 A; Kapp=0.25M) of Hsp27 is needed to signi ficantly reduce actin assembly. These findings indicate that Hsp27 needs the barbed ends of actin filaments capped for maximum sequestering activity. This could mean that in the cell, Hsp27 works intimately with capping proteins such as gelsolin, CapZ and CapG. This ma y be very relevant in the cell because most of the barbed ends of actin filame nts are capped at any given time (> 90%), allowing Hsp27 to be a major actin monomer sequestering protein. I have also found Hsp27 to be a major component of both the HeLa cell and neutrophil proteome. Hsp27 was found to be 0.07% (0.13M) of the total protein in human neutrophils and 0.4% (0.75M) in HeLa cells. These concentrations of Hsp27 exceed the Kapp for both monomer sequestration and filament stabilization.
88 I also found that Hsp27 needs to be pre-in cubated with G-actin for at least 3 min to obtain maximum monomer sequester ing. This suggests a slow initial rate of binding for Hsp27 and actin monomers. It may be possible that Hsp27 octomers bind the first actin monomer slowly, and binding could result in a conformational change in the octomer that allows additional monomer binding to occur at rapi d rate. Another possibility may be that Hsp27 prefers to bind filaments, unless it already has actin mono mers bound. In the case of monomer sequestering assays Hsp27 would be binding the gelsolin:actin seeds preferenti ally instead of the monomers. This would help explain why I see no reduction in actin assembly when no pre-incubation step is performed. In conclusion, I have demonstrated that anthrax LT can block neut rophil chemotaxis and actin assembly through cleavage of MAPKK-3/6. Th is cleavage results in a marked reduction of phospho-Hsp27, which effectively raises the amount of monomer sequestering protein in the cell. I have found non-phosphorylated Hsp27 to be an actin monomer sequestering protein whose activity is controlled by phos phorylation and enhanced by barbed end capping. Hsp27 localizes to the leading edge of neut rophils and actin tails of Listeria delivering actin monomers to zones of polymerization, where phosphorylation by MAPK AP-2 results in release of monomers. Anthrax takes advantage of the p38 MAPK pathway to effectively impair neutrophil recruitment to sites of infection by altering the equ ilibrium of phosphorylated and non-phosphorylated Hsp27, allowing excessive monomer sequestering to o ccur in the cell. This would be expected to inhibit actin polymerization, such as that seen in neutrophils treated with LT.
89 Figure 5-1. Diagrammatic representation of Hsp27 function. Hsp27, when not phophorylated has the ability to sequester actin monomers . Upon stimulation of the p38 MAPK pathway, Hsp27 is phosphorylated by MAPKAP2/3 and re leases the actin monomers. The actin monomers are then available to be ad ded on the barbed end of growing actin filaments. A phosphatase removes the phosphates from Hsp27, thus completing the cycle. LT can block phosphorylation of Hsp27 by proteolyti cally inactivating MAPKK3/6, and blocking the phosphoryla tion cycle causing an increase of nonphosphorylated Hsp27.
90 CHAPTER 6 FUTURE WORK The Effects of Edema Toxin on Neutrophils In the present study I analyzed the effects of anthrax LT on neutrophil chemotaxis and actin assembly. It would be of interest to s ee if anthrax ET has any e ffect on neutrophil actin assembly. It would also be of in terest to try both ET and LT at the same time to see if neutrophil actin assembly can be further inhibited by the presence of both toxins . This condition would mimic that in the infected host where both toxins are present. Edema toxin has also been shown to inhibit phagocytosis, which is an actin mediated process, so it is likely that ET can have an effect on the actin cytoskeleton . These e xperiments could be simply carried out by stimulating neutrophils with FMLP that have been treated with ET or ET + LT and measuring neutrophil velocities. If veloci ties were reduced, this would wa rrant a study of neutrophil actin assembly by phalloidin staining. The Effect of LT Macrophages, Dendritic Cells and Platelets I have shown that LT blocks phosphorylation of Hsp27 in neutrophils and HeLa cells, but I did not carry out any experiments to see if LT could block the phosphorylation of Hsp27 in macrophages, dendritic cells or platelets. These are three cell types that require dynamic actin remodeling in response to external stimuli in order to function properly. It is likely that anthrax LT has the same effect on Hsp27 in these other ce ll types, thus reducing their capacity for actin assembly. Macrophages must undergo chemotaxis to areas of infection, in a manner similar to that of neutrophils. Once the macrophages encoun ter foreign pathogens ph agocytosis must occur to limit spread of the infec tion. Both chemotaxis and pahgocytosis require dynamic actin remodeling. Any disruption of Hsp27 signali ng may inhibit macrophage chemotaxis and phagocytosis, thereby reducing i nnate immune function.
91 Dendritic cells must return to the lymphatic s once an antigen is en countered, so that a proper helper T-cell response can be mediated. Any impairment in actin assembly would severely disrupt this process. Dendritic cells also undergo extensive membrane ruffling when scavenging for antigen. Membrane ruffling is a di rect consequence of actin remodeling, and any defect in the actin cytoskelet on could potentially limit antigen scavenging. Antigen presentation requires that dendritic cells phagocytose antige n so that it can be processed via the MHC-II pathway. Phagocytosis requires actin assembly and membrane ruffling, which could potentially be disrupted by LT. Anthrax infection causes extensive hemorrhag ing, indicating that platelet function may be compromised in the host. Pl atelets must undergo extensive sp reading over a wound to block blood leakage from the vasculature. This spreadi ng is a consequence of extensive actin assembly that must occur for clot formation. Lethal toxin may be able to block this process by altering p38 MAPK signaling, thus blocking no rmal platelet function. Previ ous work has shown that LT blocks Erk and p38 activation in platelets, but Hsp27 phosphorylation status has not been studied . Platelet aggregation was recently f ound to be impaired after LT treatment . Role of Hsp27 and MAPK in Listeria and Neutrophil Motility I proposed that Hsp27 is deliv ering monomers to polymeri zation zones when it is not phosphorylated, but direct evidence for this is la cking, although my data seems to support this theory. It would be nice to determine if p38 MAPK and MAPKAP-2 localize to Listeria actin tails and the leading edge of ne utrophils. This would provide fu rther evidence for my hypothesis because p38 and MAPAKP-2 activation are re quired for Hsp27 phosphorylation. Specifically it would be nice to see if the phosphorylated forms of these proteins are present in zones of active polymerization. It would also be interesting to knock down Hs p27 is HL-60 cells that were differentiated to neutrophils, to see what effect this would have on motility. Another interesting
92 experiment would be to transfect cells with a constitutively active form of MAPAKAP-2, to see if this enhanced or slowed Listeria motility. A constitutively activ e form of MAPKAP-2 would also be expected to block the effects of LT. The same experiment could be performed by transfecting Hsp27 EE or AA into HeLa cells to see what effect this would have on Listeria motility. To date these kinases have never been implicated in intracellular Listeria motility. It is of interest to note that the PH doma in of Akt does loca lize to intracellular Listeria , implying that Akt may have a role in Listeria motility . If Akt does localize to Listeria actin tails, then it may be responsible for phosphorylating Hsp27 on se rine 82, causing it to release monomers for actin polymerization. Localization of Akt coul d be confirmed by generating GFP-Akt (green flurosecent protein-Akt) constructs . These constructs could be trasfected into HeLa cells before Listeria infection. A constitutively active Akt could be generated to see if Hsp27 phosphorylation is enhanced. I would exp ect there to be very little Hsp27 in Listeria actin tails if Akt does phosphorylate Hsp27, because I have shown that phosphorylated Hsp27 does not localize to Listeria actin tails. Mutational Analysis of Hsp27 Hsp27 can be phosphorylated on serines 15, 78 a nd 82, but a detailed kinetic analysis in which each these serines are individually mu tated to alanines (non-phosphorylated) or glutamines (pseudo-phosphorylated) has not been perf ormed. It would be of interest to see how each of the mutant proteins inte ract with actin, speci fically if the mutant proteins can still sequester monomers. For instance, changing seri ne 15 to glutamic acid may abolish monomer sequestering activity. This would indicate serine 15 phosphorylation controls monomer sequestering. Mutating these residues in combinati ons would also be of in terest to see if these mutations have any effect on actin binding. Puri fied proteins could al so be expressed in Listeria
93 infected HeLa cells to see if any particular residue plays a role in targeting Hsp27 to Listeria actin tails.
94 LIST OF REFERENCES 1. Health Aspects of Chemical and Biological Weapons. Report of a Consultant Working Group. Geneva: World Health Organization;1970 2. Jernigan DB, Raghunathan PL, Bell BP, Brechne r R, Bresnitz EA, Butler JC, Cetron M, Cohen M, Doyle T, Fischer M, Greene C, Griffith KS, Guarner J, Hadler JL, Hayslett JA, Meyer R, Petersen LR, Phillips M, Pinner R, Popovic T, Quinn CP, Reefhuis J, Reissman D, Rosenstein N, Schuchat A, Shieh WJ, Siegal L, Swerdlow DL, Tenover FC, Traeger M, Ward JW, Weisfuse I, Wiersma S, Yeskey K, Zaki S, Ashford DA, Perkins BA, Ostroff S, Hughes J, Fleming D, Koplan JP and Gerberding JL. Investigation of bioterrorism-related anthrax, United States, 2001: epidemiologic findings. Emerg Infect Dis 2002;8:1019-28 3. Keim P, Smith KL. Bacillus anthracis evolution and epidemiology. Curr Top Microbiol Immunol 2002;271:21-32 4. Keppie J, Smith H and Harris-Smith PW. The chemical basis of the virulence of bacillus anthracis. II. Some biological properties of bacterial products . Br J Exp Pathol 1953;34:486-96 5. Smith H, Stoner HB. Anthrax toxic complex. Fed Proc 1967;26:1554-7 6. Okinaka RT, Cloud K, Hampton O, Hoffmaster AR, Hill KK, Keim P, Koehler TM, Lamke G, Kumano S, Mahillon J, Manter D, Martinez Y, Ricke D, Svensson R and Jackson PJ. Sequence and organization of pXO1, the large Bacillus anthracis plasmid harboring the anthrax toxin genes. J Bacteriol 1999;181:6509-15 7. Green BD, Battisti L, Koehler TM, Thorne CB and Ivins BE. Demonstration of a capsule plasmid in Bacillus anthracis. Infect Immun 1985;49:291-7 8. Read TD, Peterson SN, Tourasse N, et al. Th e genome sequence of Bacillus anthracis Ames and comparison to closely related bacteria. Nature 2003;423:81-6 9. Spencer RC. Bacillus anthracis. J Clin Pathol 2003;56:182-7 10. Friedlander AM. Macrophages are sensitive to anthrax lethal toxin through an aciddependent process. J Biol Chem 1986;261:7123-6 11. Bradley KA, Mogridge J, Mourez M, Collier RJ and Young JA. Identification of the cellular receptor for anthrax toxin. Nature 2001;414:225-9 12. Scobie HM, Rainey GJ, Bradley KA and Yo ung JA. Human capillary morphogenesis protein 2 functions as an anthrax toxin receptor. Proc Natl Acad Sci U S A 2003;100:5170-4
95 13. Gordon VM, Klimpel KR, Arora N, Henderson MA and Leppla SH. Proteolytic activation of bacterial toxins by eukaryotic cells is perfor med by furin and by additional cellular proteases. Infect Immun 1995;63:82-7 14. Mogridge J, Cunningham K, Lacy DB, Mour ez M and Collier RJ. The lethal and edema factors of anthrax toxin bind only to oligomeric forms of the pr otective antigen. Proc Natl Acad Sci U S A 2002;99:7045-8 15. Mogridge J, Cunningham K and Collier RJ. St oichiometry of anthrax toxin complexes. Biochemistry 2002;41:1079-82 16. Duesbery NS, Webb CP, Leppla SH, et al. Pr oteolytic inactivation of MAP-kinase-kinase by anthrax lethal factor . Science 1998;280:734-7 17. Leppla SH. Anthrax toxin edema factor: a bacter ial adenylate cyclase th at increases cyclic AMP concentrations of eukaryotic ce lls. Proc Natl Acad Sci USA 1982;79:3162-6 18. O'Brien J, Friedlander A, Dreier T, Ezze ll J and Leppla S. Effects of anthrax toxin components on human neutrophils . Infect Immun 1985;47:306-10 19. Roux PP, Blenis J. ERK and p38 MAPK-activat ed protein kinases: a family of protein kinases with diverse biological func tions. Microbiol Mol Biol Rev 2004;68:320 20. Crespo P, Xu N, Simonds WF and Gutkind JS. Ras-dependent activation of MAP kinase pathway mediated by G-protein beta gamma subunits. Nature 1994;369:418-20 21. Faure M, Voyno-Yasenetskaya TA and Bour ne HR. cAMP and beta gamma subunits of heterotrimeric G proteins stimulate the mitogenactivated protein kinase pathway in COS-7 cells. J Biol Chem 1994;269:7851-4 22. Anderson NG, Maller JL, Tonks NK and Sturgill TW. Requirement for integration of signals from two distinct phosphorylation pathways fo r activation of MAP kinase. Nature 1990;343:6513 23. Dent P, Haser W, Haystead TA, Vincent LA , Roberts TM and Sturgill TW. Activation of mitogen-activated protein kinase kinase by vRaf in NIH 3T3 cells and in vitro. Science 1992;257:1404-7 24. Force T, Bonventre JV, Heidecker G, Rapp U, Avruch J and Kyriakis JM. Enzymatic characteristics of the c-Raf1 protein kinase. Proc Natl Acad Sci U S A 1994;91:1270-4 25. Kyriakis JM, App H, Zhang XF, Banerjee P, Brautigan DL, Rapp UR and Avruch J. Raf-1 activates MAP kinase-kin ase. Nature 1992;358:417-21
96 26. Rousseau S, Houle F, Landry J and Huot J. p38 MAP kinase activation by vascular endothelial growth factor mediat es actin reorganization and cell migration in human endothelial cells. Oncogene 1997;15:2169-77 27. Hale KK, Trollinger D, Rihanek M and Mant hey CL. Differential ex pression and activation of p38 mitogen-activated protein kinase alpha, beta, gamma, and delta in inflammatory cell lineages. J Immunol 1999;162:4246-52 28. Raingeaud J, Gupta S, Rogers JS, Dicken s M, Han J, Ulevitch RJ and Davis RJ. Proinflammatory cytokines and environmental stre ss cause p38 mitogen-activated protein kinase activation by dual phosphorylation on tyrosi ne and threonine. J Biol Chem 1995;270:7420-6 29. Hannigan M, Zhan L, Ai Y and Huang CK. The role of p38 MAP kinase in TGF-beta1induced signal transduction in human neut rophils. Biochem Biophys Res Commun 1998;246:558 30. Freshney NW, Rawlinson L, Guesdon F, Jones E, Cowley S, Hsuan J and Saklatvala J. Interleukin-1 activates a novel pr otein kinase cascade that resu lts in the phosphorylation of Hsp27. Cell 1994;78:1039-49 31. Wesselborg S, Bauer MK, Vogt M, Schmitz ML and Schulze-Osthoff K. Activation of transcription factor NF-kappaB a nd p38 mitogen-activated protein ki nase is mediated by distinct and separate stress effector pathways. J Biol Chem 1997;272:12422-9 32. Kim KY, Kim BC, Xu Z and Kim SJ. Mixed lineage kinase 3 (MLK3)-activated p38 MAP kinase mediates transforming growth factor-bet a-induced apoptosis in hepatoma cells. J Biol Chem 2004;279:29478-84 33. Keesler GA, Bray J, Hunt J, Johnson DA, Gl eason T, Yao Z, Wang SW, Parker C, Yamane H, Cole C and Lichenstein HS. Purification a nd activation of recombinant p38 isoforms alpha, beta, gamma, and delta. Pr otein Expr Purif 1998;14:221-8 34. Briata P, Forcales SV, Ponassi M, Corte G, Chen CY, Karin M, Puri PL and Gherzi R. p38dependent phosphorylation of the mRNA decay-promo ting factor KSRP controls the stability of select myogenic transcri pts. Mol Cell 2005;20:891-903 35. Hitti E, Iakovleva T, Brook M, Deppenmeier S, Gruber AD, Radzioch D, Clark AR, Blackshear PJ, Kotlyarov A and Gaestel M. Mitoge n-activated protein kina se-activated protein kinase 2 regulates tumor necrosis factor mRNA stability and translation mainly by altering tristetraprolin expression, stabil ity, and binding to adenine/uridin e-rich element. Mol Cell Biol 2006;26:2399-407 36. Dean JL, Sully G, Clark AR and Saklatvala J. The involvement of AU-rich element-binding proteins in p38 mitogen-activat ed protein kinase pathway-medi ated mRNA stabilisation. Cell Signal 2004;16:1113-21
97 37. Tan Y, Rouse J, Zhang A, Cariati S, Cohen P and Comb MJ. FGF and stress regulate CREB and ATF-1 via a pathway involving p38 MAP kinase and MAPKAP kinase-2. Embo J 1996;15:4629-42 38. McLaughlin MM, Kumar S, McDonnell PC, Van Horn S, Lee JC, Livi GP and Young PR. Identification of mitogen-activat ed protein (MAP) kina se-activated protei n kinase-3, a novel substrate of CSBP p38 MAP ki nase. J Biol Chem 1996;271:8488-92 39. Rouse J, Cohen P, Trigon S, Morange M, Al onso-Llamazares A, Zamanillo D, Hunt T and Nebreda AR. A novel kinase cascade triggered by stress and heat shock that stimulates MAPKAP kinase-2 and phosphoryl ation of the small heat shock proteins. Cell 1994;78:1027-37 40. Park JM, Greten FR, Li ZW and Karin M. Macrophage apoptosis by an thrax lethal factor through p38 MAP kinase inhi bition. Science 2002;297:2048-51 41. Fukao T. Immune system paralysis by anthrax le thal toxin: the roles of innate and adaptive immunity. Lancet In fect Dis 2004;4:166-70 42. Agrawal A, Lingappa J, Leppla SH, et al. Impairment of dendritic cells and adaptive immunity by anthrax lethal toxin. Nature 2003;424:329-34 43. Comer JE, Chopra AK, Peterson JW and Koni g R. Direct inhibition of T-lymphocyte activation by anthrax toxins in vivo. Infect Immun 2005;73:8275-81 44. English D. The Remarkable Neutrophil! In: Gabrilovich ID . The Neutrophils: New Outlook for Old Cells. Imperial Co llege Press, 2nd ed. 2005: 1-34 45. Anastassova-Kristeva M. The origin and devel opment of the immune system with a view to stem cell therapy. J Hematother Stem Cell Res 2003;12:137-54 46. Moore MA. Commentary: the role of cell migration in the ont ogeny of the lymphoid system. Stem Cells Dev 2004;13:1-21 47. Semerad CL, Liu F, Gregory AD, Stumpf K and Link DC. G-CSF is an essential regulator of neutrophil trafficking from the bone marrow to the blood. Immunity 2002;17:413-23 48. Cheers C, Haigh AM, Kelso A, Metcalf D, Stanley ER and Young AM. Production of colony-stimulating factors (CSFs) during infection: separate de terminations of macrophage-, granulocyte-, granulocyte-macrophage-, and multi-CSFs. Infect Immun 1988;56:247-51 49. Metcalf D. The role of the colony-stimulatin g factors in resistance to acute infections. Immunol Cell Biol 1987;65 ( Pt 1):35-43 50. Cairo MS, Mauss D, Kommared dy S, Norris K, van de Ven C and Modanlou H. Prophylactic or simultaneous administration of recombinant hu man granulocyte colony s timulating factor in the treatment of group B streptococcal sepsis in neonatal rats. Pe diatr Res 1990;27:612-6
98 51. Yasuda H, Ajiki Y, Shimozato T, Kasaha ra M, Kawada H, Iwata M and Shimizu K. Therapeutic efficacy of granulocyte colony-stimul ating factor alone and in combination with antibiotics against Pseudomonas aeruginosa in fections in mice. Infect Immun 1990;58:2502-9 52. Boogaerts M, Cavalli F, Cortes-Funes H, Ga tell JM, Gianni AM, Khayat D, Levy Y and Link H. Granulocyte growth factors: achie ving a consensus. Ann Oncol 1995;6:237-44 53. Mock BH, Schauwecker DS, English D, Y oung KA and Wellman HN. In vivo kinetics of canine leukocytes labeled with technetium-99 m HM-PAO and indium-111 tropolonate. J Nucl Med 1988;29:1246-51 54. Hayashi F, Means TK and Luster AD. To ll-like receptors stimulate human neutrophil function. Blood 2003;102:2660-9 55. Suttmann H, Lehan N, Bohle A and Brandau S. Stimulation of neutrophil granulocytes with Mycobacterium bovis bacillus Calmette-Gueri n induces changes in phenotype and gene expression and inhibits spontaneous apoptosis. Infect Immun 2003;71:4647-56 56. Ancelin M, Chollet-Martin S, Herve MA, Le grand C, El Benna J and Perrot-Applanat M. Vascular endothelial growth factor VEGF189 induces human neutrophil chemotaxis in extravascular tissue via an au tocrine amplification mechan ism. Lab Invest 2004;84:502-12 57. Grenier A, Combaux D, Chastre J, Gouge rot-Pocidalo MA, Gibert C, Dehoux M and Chollet-Martin S. Oncostatin M production by blood and alveolar neutrophils during acute lung injury. Lab Invest 2001;81:133-41 58. Topham MK, Carveth HJ, McIntyre TM, Prescott SM and Zimmerman GA. Human endothelial cells regulate polymorphonuclear leukocyte degr anulation. Faseb J 1998;12:733-46 59. Cassatella MA, Guasparri I, Ceska M, Bazzoni F and Rossi F. Interferon-gamma inhibits interleukin-8 production by human polymor phonuclear leucocytes. Immunology 1993;78:177-84 60. Koga Y, Matsuzaki A, Suminoe A, Hattori H and Hara T. Neutrophil-derived TNF-related apoptosis-inducing ligand (TRAIL): a novel mechanism of antitumor effect by neutrophils. Cancer Res 2004;64:1037-43 61. Vulcano M, Alves Rosa MF, Minnucci FS, Chernavsky AC and Isturiz MA. N-formylmethionyl-leucyl-phenylalanine (fMLP) inhibits tumour necrosis fact or-alpha (TNF-alpha) production on lipopolysaccharide (LPS)-stimulat ed human neutrophils. Clin Exp Immunol 1998;113:39-47 62. Hatano Y, Kasama T, Iwabuchi H, Hanaoka R, Takeuchi HT, Jing L, Mori Y, Kobayashi K, Negishi M, Ide H and Adachi M. Macrophage in flammatory protein 1 alpha expression by synovial fluid neutrophils in rheumato id arthritis. Ann Rheum Dis 1999;58:297-302
99 63. Kunkel SL, Lukacs N and Strieter RM. Expre ssion and biology of neut rophil and endothelial cell-derived chemokines. Semin Cell Biol 1995;6:327-36 64. Baggiolini M. Chemokines in pathology and medicine. J Intern Med 2001;250:91-104 65. Xing L, Remick DG. Relative cytokine and cytokine inhibitor pr oduction by mononuclear cells and neutrophils. Shock 2003;20:10-6 66. Ethuin F, Delarche C, Gougerot-Pocidalo MA , Eurin B, Jacob L and Chollet-Martin S. Regulation of interleukin 12 p40 and p70 producti on by blood and alveolar phagocytes during severe sepsis. Lab Invest 2003;83:1353-60 67. Brandt ER, Linnane AW and Devenish RJ. E xpression of IFN A genes in subpopulations of peripheral blood cells. Br J Haematol 1994;86:717-25 68. Keel M, Ungethum U, Steckholzer U, Nieder er E, Hartung T, Trentz O and Ertel W. Interleukin-10 counterregulates proinflammatory cytokine-induced inhibition of neutrophil apoptosis during severe sepsis. Blood 1997;90:3356-63 69. Movat HZ, Burrowes CE, Cybulsky MI and Dinarello CA. Acute inflammation and a Shwartzman-like reaction induced by interleukin1 and tumor necrosis fa ctor. Synergistic action of the cytokines in the induc tion of inflammation and microva scular injury. Am J Pathol 1987;129:463-76 70. Pace JL, Russell SW, Schreiber RD, Altman A and Katz DH. Macrophage activation: priming activity from a T-cell hybridoma is attribut able to interferon-ga mma. Proc Natl Acad Sci U S A 1983;80:3782-6 71. Ward PA, Newman LJ. A neutrophil chemot actic factor from human C'5. J Immunol 1969;102:93-9 72. Charo IF, Yuen C, Perez HD and Goldstein IM . Chemotactic peptides modulate adherence of human polymorphonuclear leukocytes to monolayers of cultured endothelial cells. J Immunol 1986;136:3412-9 73. Christman JW, Petras SF, Vacek PM and Davi s GS. Rat alveolar macrophage production of chemoattractants for neutrophils: response to Escherichia coli endotoxin. Infect Immun 1989;57:810-6 74. Bainton DF. Sequential degranulation of th e two types of polymor phonuclear leukocyte granules during phagocytosis of micr oorganisms. J Cell Biol 1973;58:249-64
100 75. Abbassi O, Kishimoto TK, McIntire LV and Smith CW. Neutrophil adhesion to endothelial cells. Blood Cells 1993;19: 245-59; discussion 259-60 76. Lomakina EB, Waugh RE. Dynamics of incr eased neutrophil adhesion to icam-1 after contacting immobilized IL-8. A nn Biomed Eng 2006;34:1553-63 77. Marlin SD, Springer TA. Purifi ed intercellular adhesion molecule-1 (ICAM-1) is a ligand for lymphocyte function-associated antigen 1 (LFA-1). Cell 1987;51:813-9 78. Lane TA, Lamkin GE and Wancewicz E. Mo dulation of endothelial cell expression of intercellular adhesion molecule 1 by protein kinase C activation. Bi ochem Biophys Res Commun 1989;161:945-52 79.Thomas KM, Pyun HY and Navarro J. Molecular cloning of the fMet-Leu-Phe receptor from neutrophils. J Biol Chem 1990;265:20061-4 80. Williams LT, Snyderman R, Pike MC and Lefkowitz RJ. Specific receptor sites for chemotactic peptides on human polymorphonuclear leukocytes. Proc Natl Acad Sci U S A 1977;74:1204-8 81. Northup JK, Sternweis PC, Smigel MD, Schlei fer LS, Ross EM and Gilman AG. Purification of the regulatory component of adenylate cy clase. Proc Natl Acad Sci U S A 1980;77:6516-20 82. Park D, Jhon DY, Lee CW, Lee KH and Rhee SG. Activation of phospholipase C isozymes by G protein beta gamma subunits . J Biol Chem 1993;268:4573-6 83. Jin T, Zhang N, Long Y, Parent CA and Devreotes PN. Localization of the G protein betagamma complex in living cells during chemotaxis. Science 2000;287:1034-6 84. Krugmann S, Hawkins PT, Pryer N and Brasel mann S. Characterizing the interactions between the two subunits of the p101/p110gamma phosphoi nositide 3-kinase and their role in the activation of this enzyme by G beta gamma subunits. J Biol Chem 1999;274:17152-8 85. Stephens L, Smrcka A, Cooke FT, Jacks on TR, Sternweis PC and Hawkins PT. A novel phosphoinositide 3 kinase activity in myeloid-derive d cells is activated by G protein beta gamma subunits. Cell 1994;77:83-93 86. Han J, Luby-Phelps K, Das B, Shu X, Xia Y, Mosteller RD, Krishna UM, Falck JR, White MA and Broek D. Role of substr ates and products of PI 3-kinase in regulating activation of Racrelated guanosine triphosphata ses by Vav. Science 1998;279:558-60 87. Higgs HN, Pollard TD. Activation by Cdc42 and PIP(2) of Wiskott-Aldrich syndrome protein (WASp) stimulates actin nucleati on by Arp2/3 complex. J Cell Biol 2000;150:1311-20
101 88. Zigmond SH, Joyce M, Yang C, Brown K, Huang M and Pring M. Mechanism of Cdc42induced actin polymerization in neut rophil extracts. J Cell Biol 1998;142:1001-12 89. Wegner A, Isenberg G. 12-fold difference betw een the critical monome r concentrations of the two ends of actin filaments in physiologi cal salt conditions. Proc Natl Acad Sci U S A 1983;80:4922-5 90. Arisaka F, Noda H and Maruyama K. Kinetic an alysis of the polymerization process of actin. Biochim Biophys Acta 1975;400:263-74 91. Wegner A. The mechanism of ATP hydrolysis by polymer actin. Biophys Chem 1977;7:51-8 92. Kang F, Purich DL and Southwick FS. Profilin promotes barbed-end actin filament assembly without lowering the critical concen tration. J Biol Chem 1999;274:36963-72 93. Wolven AK, Belmont LD, Mahoney NM, Almo SC and Drubin DG. In vivo importance of actin nucleotide exchange catalyzed by profilin. J Cell Biol 2000;150:895-904 94. De La Cruz EM, Ostap EM, Brundage RA, Re ddy KS, Sweeney HL and Safer D. Thymosinbeta(4) changes the conformation and dynamics of actin monomers. Biophys J 2000;78:2516-27 95. Safer D, Sosnick TR and Elzinga M. Thymosin beta 4 binds actin in an extended conformation and contacts both the barbed and pointed ends. Biochemistry 1997;36:5806-16 96. Nyman T, Page R, Schutt CE, Karlsson R an d Lindberg U. A crosslinked profilin-actin heterodimer interferes with elongation at the fast-growing en d of F-actin. J Biol Chem 2002;277:15828-33 97. Yu FX, Johnston PA, Sudhof TC and Yin HL. gCap39, a calcium ionand polyphosphoinositide-regulated actin capping protein. Science 1990;250:1413-5 98. McGrath JL, Osborn EA, Tardy YS, Dewey CF, Jr. and Hartwig JH. Regulation of the actin cycle in vivo by actin filament severi ng. Proc Natl Acad Sci U S A 2000;97:6532-7 99. Lin KM, Mejillano M and Yin HL. Ca2+ regulation of gelsolin by its C-terminal tail. J Biol Chem 2000;275:27746-52 100. Janmey PA, Stossel TP. Modulation of ge lsolin function by phosphatidylinositol 4,5bisphosphate. Nature 1987;325:362-4 101. Mullins RD, Heuser JA and Pollard TD. Th e interaction of Arp2/ 3 complex with actin: nucleation, high affinity pointed end capping, and formation of branching networks of filaments. Proc Natl Acad Sci U S A 1998;95:6181-6 102. Mullins RD, Pollard TD. Rho-family GTPa ses require the Arp2/3 complex to stimulate actin polymerization in Acanthamoeba extracts. Curr Biol 1999;9:405-15
102 103. Ridley AJ, Hall A. The small GTP-binding pr otein rho regulates the assembly of focal adhesions and actin stress fi bers in response to growth factors. Cell 1992;70:389-99 104. Ridley AJ, Paterson HF, Johnston CL, Di ekmann D and Hall A. The small GTP-binding protein rac regulates growth factor-induced membrane ruffling. Cell 1992;70:401-10 105. Castellano F, Montcourrier P, Guillemot JC, Gouin E, Machesky L, Cossart P and Chavrier P. Inducible recruitment of Cdc42 or WASP to a cell-surface receptor triggers actin polymerization and filopodium formation. Curr Biol 1999;9:351-60 106. Machesky LM, Mullins RD, Higgs HN, Kaiser DA, Blanchoin L, May RC, Hall ME and Pollard TD. Scar, a WASp-related protein, activates nucleation of actin filaments by the Arp2/3 complex. Proc Natl Acad Sci U S A 1999;96:3739-44 107. Dabiri GA, Sanger JM, Portnoy DA, Southw ick FS. Listeria monocytogenes moves rapidly through the host-cell cytoplasm by inducing directional actin assembly. Proc Natl Acad Sci USA 1990; 87 6068-72 108. Mengaud J, Ohayon H, Gounon P, Mege RM and Cossart P. E-cadherin is the receptor for internalin, a surface protein requ ired for entry of L. monocytoge nes into epithelial cells. Cell 1996;84:923-32 109. Shen Y, Naujokas M, Park M and Ireton K. In IB-dependent internalization of Listeria is mediated by the Met receptor ty rosine kinase. Cell 2000;103:501-10 110. Lecuit M, Hurme R, Pizarro-Cerda J, Ohayon H, Geiger B and Cossart P. A role for alphaand beta-catenins in bact erial uptake. Proc Natl Acad Sci U S A 2000;97:10008-13 111. Tang P, Sutherland CL, Gold MR and Finl ay BB. Listeria monocytogenes invasion of epithelial cells requires the MEK-1/ERK-2 mitoge n-activated protein kinase pathway. Infect Immun 1998;66:1106-12 112. Ireton K, Payrastre B, Chap H, Ogawa W, Sa kaue H, Kasuga M and Cossart P. A role for phosphoinositide 3-kinase in bacterial invasion. Science 1996;274:780-2 113. Bierne H, Miki H, Innocenti M, Scita G, Gertler FB, Take nawa T and Cossart P. WASPrelated proteins, Abi1 and Ena/VASP are requir ed for Listeria invasion induced by the Met receptor. J Cell Sci 2005;118:1537-47 114. Schuerch DW, Wilson-Kubalek EM and Tweten RK. Molecular basis of listeriolysin O pH dependence. Proc Natl Acad Sci U S A 2005;102:12537-42 115. Kocks C, Gouin E, Tabouret M, Berche P, Ohayon H and Cossart P. L. monocytogenesinduced actin assembly requires the actA ge ne product, a surface protein. Cell 1992;68:521-31
103 116. Smith GA, Theriot JA and Portnoy DA. The tandem repeat domain in the Listeria monocytogenes ActA protein controls the rate of actin-based motility, the percentage of moving bacteria, and the localization of vasodilator-s timulated phosphoprotein and profilin. J Cell Biol 1996;135:647-60 117. Loisel TP, Boujemaa R, Pantaloni D and Carl ier MF. Reconstitution of actin-based motility of Listeria and Shigella using pure proteins. Nature 1999;401:613-6 118. Quinn CP, Shone CC, Turnbull PC and Melling J. Purification of anthrax-toxin components by high-performance anion-exchange, ge l-filtration and hydr ophobic-interaction chromatography. Biochem J 1988;252:753-8 119. Shalit M, Dabiri GA and Southwick FS. Plat elet-activating factor both stimulates and "primes" human polymorphonuclear leukocyte ac tin filament assembly. Blood 1987;70:1921-7 120. Amrein PC, Stossel TP. Prevention of de gradation of human polymorphonuclear leukocyte proteins by diisopropylfluor ophosphate. Blood 1980;56:442-7 121. Allgaier B, Shi M, Luo D and Koenig JM. Spontaneous and Fas-mediated apoptosis are diminished in umbilical cord blood neutrophils compared with adult neutrophils. J Leukoc Biol 1998;64:331-6 122. Zeile WL, Purich DL and Southwick FS. Recognition of two classes of oligoproline sequences in profilin-mediated acceleration of actin-based Shigella motility. J Cell Biol 1996;133:49-59 123. Wallace PJ, Wersto RP, Packman CH and Lichtman MA. Chemotactic peptide-induced changes in neutrophil actin conf ormation. J Cell Biol 1984;99:1060-5 124. Downey GP, Butler JR, Tapper H, et al. Im portance of MEK in neutrophil microbicidal responsiveness. J Immunol 1998;160:434-43 125. Southwick FS, Dabiri GA and Stossel TP. Neut rophil actin dysfunction is a genetic disorder associated with partial impairment of neutrophil ac tin assembly in three family members. J Clin Invest 1988;82:1525-31 126. Berkelman T, Stenstedt T. 2-D Electrophor esis Using Immobolized PH Gradients Principles and Methods. Amersh am Pharmacia Biotech, 1998:1-50 127. Lambert H, Charette SJ, Bernier AF, Gu imond A and Landry J. HSP27 multimerization mediated by phosphorylation-sensitive intermolecular interactions at the amino terminus. J Biol Chem 1999;274:9378-85 128. Egile C, Loisel TP, Laurent V, et al. Ac tivation of the CDC42 effector N-WASP by the Shigella flexneri IcsA protein promotes actin nucleation by Arp2/3 complex and bacterial actinbased motility. J Cell Biol 1999;146:1319-32
104 129. Young CL, Southwick FS and Weber A. Kineti cs of the interactio n of a 41-kilodalton macrophage capping protein with actin: promotio n of nucleation during prolongation of the lag period. Biochemistry 1990;29:2232-40 130. Kouyama T, Mihashi K. Fluorimetry study of N-(1-pyrenyl)iodoacetamide-labelled F-actin. Local structural change of actin protomer both on polymerization and on binding of heavy meromyosin. Eur J Biochem 1981;114:33-8 132. Lin DC, Lin S. High affinity binding of [3H]dihydrocytochala sin B to peripheral membrane proteins related to the contro l of cell shape in the human re d cell. J Biol Chem 1978;253:1415-9 133. Alberts B, Johnson A, Lewis J, Raff M, Robe rts K, Walter P. Molecular Biology of The Cell. 4th ed. Garland Science, 2002:831-905 134. Elbaum D, Mimms LT and Branton D. Modulat ion of actin polymerization by the spectrinband 4.1 complex. Biochemistry 1984;23:4813-6 135. Southwick FS, DiNubile MJ. Rabbit alveol ar macrophages contain a Ca2+-sensitive, 41,000-dalton protein which reversibly blocks the "barbed" ends of actin filaments but does not sever them. J Biol Chem 1986;261:14191-5 136. Bubb MR, Yarmola EG, Gibson BG and S outhwick FS. Depolymerization of actin filaments by profilin. Effects of profilin on capping protein function. J Biol Chem 2003;278:24629-35 137. Jernigan JA, Stephens DS, Ashford DA, et al . Bioterrorism-related inhalational anthrax: the first 10 cases reported in the United States. Emerg Infect Dis 2001;7:933-44 138. Wade BH, Wright GG, Hewlett EL, Leppla SH and Mandell GL. Anthrax toxin components stimulate chemotaxis of human polymorphonucl ear neutrophils. Proc Soc Exp Biol Med 1985;179:159-62 139. Crawford MA, Aylott CV, Bourdeau RW and Bokoch GM. Bacillus anthracis toxins inhibit human neutrophil NADPH oxidase activity. J Immunol 2006;176:7557-65 140. Howard TH, Oresajo CO. The kinetics of ch emotactic peptide-induced change in F-actin content, F-actin distribution, and the shap e of neutrophils. J Cell Biol 1985;101:1078-85 141. Szczur K, Xu H, Atkinson S, Zheng Y a nd Filippi MD. Rho GTPase CDC42 regulates directionality and random movement via distin ct MAPK pathways in neutrophils. Blood 2006 142. Benndorf R, Hayess K, Ryazantsev S, Wieske M, Behlke J and Lutsch G. Phosphorylation and supramolecular organization of murine sma ll heat shock protein HSP25 abolish its actin polymerization-inhibiting activ ity. J Biol Chem 1994;269:20780-4
105 143. Gardiner EM, Pestonjamasp KN, Bohl BP, Chamberlain C, Hahn KM and Bokoch GM. Spatial and temporal analysis of Rac activation during live ne utrophil chemotaxis. Curr Biol 2002;12:2029-34 144. Liu L, Schwartz BR, Lin N, Winn RK and Harlan JM. Requirement for RhoA kinase activation in leukocyte de-a dhesion. J Immunol 2002;169:2330-6 145. Nahas N, Molski TF, Fernandez GA and Sha'afi RI. Tyrosine phosphorylation and activation of a new mitogen-activated protei n (MAP)-kinase cascade in human neutrophils stimulated with various agonists. Biochem J 1996;318 ( Pt 1):247-53 146. Waterman WH, Sha'afi RI. Effects of gran ulocyte-macrophage colony-stimulating factor and tumour necrosis factor-alpha on tyrosi ne phosphorylation and act ivation of mitogenactivated protein kinases in human neut rophils. Biochem J 1995;307 ( Pt 1):39-45 147. Janway CA, Travers P, Walport M, Shlomchik MJ. Immunobiology: The Immune System in Health and Diseas e. Garland Science, 5th ed. 2005: 37-96 148. Pichon S, Bryckaert M and Berrou E. Contro l of actin dynamics by p38 MAP kinase-Hsp27 distribution in the lamellipodium of sm ooth muscle cells. J Cell Sci 2004;117:2569-77 149. Guay J, Lambert H, Gingras-Breton G, Lavoie JN, Huot J and Landry J. Regulation of actin filament dynamics by p38 map kinase-mediated phos phorylation of heat shock protein 27. J Cell Sci 1997;110 ( Pt 3):357-68 150. Lavoie JN, Hickey E, Weber LA and Landr y J. Modulation of actin microfilament dynamics and fluid phase pinocyt osis by phosphorylation of heat shock protein 27. J Biol Chem 1993;268:24210-4 151. Gerthoffer WT, Gunst SJ. Invited review: focal adhesion and small heat shock proteins in the regulation of actin rem odeling and contractility in smooth muscle. J Appl Physiol 2001;91:963-72 152. Dorion S, Landry J. Activation of the mitoge n-activated protein kina se pathways by heat shock. Cell Stress Chaperones 2002;7:200-6 153. Landry J, Lambert H, Zhou M, et al. Huma n HSP27 is phosphorylated at serines 78 and 82 by heat shock and mitogen-activated kinases th at recognize the same amino acid motif as S6 kinase II. J Biol Chem 1992;267:794-803 154. Theriault JR, Lambert H, Chavez-Zobel AT, Ch arest G, Lavigne P and Landry J. Essential role of the NH2-terminal WD/E PF motif in the phosphorylation-act ivated protective function of mammalian Hsp27. J Biol Chem 2004;279:23463-71
106 155. Lavoie JN, Lambert H, Hickey E, Weber LA and Landry J. Modulation of cellular thermoresistance and actin filament stability accompanies phosphorylation-induced changes in the oligomeric structure of heat shoc k protein 27. Mol Cell Biol 1995;15:505-16 156. Zu YL, Ai Y, Gilchrist A, Labadia ME, Sha'afi RI and Huang CK. Activation of MAP kinase-activated protein kinase 2 in human neutrophils after phorbol ester or fMLP peptide stimulation. Blood 1996;87:5287-96 157. Rane MJ, Pan Y, Singh S, et al. Heat shock protein 27 controls apop tosis by regulating Akt activation. J Biol Chem 2003;278:27828-35 158. Rane MJ, Coxon PY, Powell DW, et al. p38 Kinase-dependent MAPKAPK-2 activation functions as 3-phosphoinositide-dependent kinase-2 for Akt in human neutrophils. J Biol Chem 2001;276:3517-23 159. Kau JH, Sun DS, Tsai WJ, et al. Antiplatelet ac tivities of anthra x lethal toxin are associated with suppressed p42/44 and p38 mitogen-activated pr otein kinase pathways in the platelets. J Infect Dis 2005;192:1465-74 160. Sidhu G, Li W, Laryngakis N, Bishai E, Balla T and Southwick F. Phosphoinositide 3kinase is required for intr acellular Listeria monocytogenes actin-based motility and filopod formation. J Biol Chem 2005;280:11379-86
107 BIOGRAPHY Russell Lavon During, Jr. was born October 5, 1980, in Tifton, GA. Russ graduated from Irwin County High School in 1999. He graduate d magna cum laude from Valdosta State University with a bachelorâ€™s degree in biological sciences in 2003. He entered the Interdisciplinary Program in Biomedical Sciences at the University of Florida in 2003, where the work presented in this dissertation was perf ormed. Russ plans on continuing his education by attending either Emory University or the University of Florida in pursuit of a masterâ€™s degree in physician assistant studies.