Studies of the effects of antimicrobial peptides on model membranes and relevant pathogens

Studies of the effects of antimicrobial peptides on model membranes and relevant pathogens Tânia Martins da Silva Dissertação de Mestrado em Bioquímica Universidade do Porto Faculdade de Ciências Instituto Ciências Biomédicas Abel Salazar 2011 Tânia Martins da Silva Studies of the effects of antimicrobial peptides on model membranes and relevant pathogens Dissertação de Candidatura ao grau de Mestre em Bioquímica da Universidade do Porto Orientador – Doutora Margarida Bastos Categoria – Professora Associada com Agregação Afiliação – CIQ(UP), Departamento Química e Bioquímica, Faculdade de Ciências da Universidade do Porto Co-Orientador – Doutora Maria Salomé Gomes Categoria – Professora Associada Afiliação – Instituto de Ciências Biomédicas Abel Salazar e Instituto de Biologia Molecular e Celular 2011 1 CONTENTS List of symbols and abbreviations…………………………………………………………… 3 Acknowledgements………………………………………………………………………….... 5 Abstract……………………………………………………………………………………….... 7 Resumo……………………………………………………………………………………...…. 9 Thesis Organization…………………………………………………………………………… 10 PART I – INTRODUCTION………………………………………………………………... 13 1. General Introduction…………………………………………………………….……….... 15 2. Pathogens………………………………………………………………………………...… 17 2.1. Leishmania…………………………………………………………………………. 17 2.1.1. Leishmaniasis………………………………………………………………… 17 2.1.2. Life cycle………………………………………..…………………………….. 19 2.1.3. Treatments……………………………………………………………………. 20 2.1.4. Leishmania as a target of AMPs…………………………………………… 21 2.2. Mycobacterium……………………………………………..……………………… 23 2.2.1. Mycobacteria cell wall……………………………………………………….. 23 2.2.2. Mycobacterium avium……………………………………………….………. 24 2.2.3. Treatments…………………………………………………………..……….. 26 2.2.4. Mycobacteria as a target of AMPs…………………………………………. 27 3. Antimicrobial peptides………………………………………………………………..……. 29 3.1. Structural characteristics……………………………………………………..…... 30 3.2. Mechanism of action………………………………………………………………. 31 3.3. Selectivity / Toxicity……………………………………………………………….. 34 3.4. AMPs in the immunity……………………………………………………………... 36 3.5. Resistance to AMPs………………………………………………………………. 37 3.6. AMPs in the clinic: Success, Problems and Challenges……………………… 38 3.7. Cecropin A-melittin derived peptides……………………………………………. 40 3.8. Lactoferrin-derived peptides……………………………………………………… 43 4. Lipids in Biological and Model Membranes……………………………………………... 47 4.1. Lipid Polymorphism…………………………………………………..…………… 47 4.2. AMP interaction with Lipid Membranes as revealed by Model Studies……… 51 PART II – OBJECTIVES…………………………………………………..……………... 53 2 PART III – EXPERIMENTAL METHODS AND RESULTS……………………..…………… 57 III.1 – Antimicrobial activity of lactoferrin-derived peptides against Leishmania... 59 1. Material and Methods………………………………..……………………………………. 61 2. Results……………………………………………..……………………………………….. 67 2.1. Determination of the antimicrobial activity of lactoferrin-derived peptides against L. donovani promastigotes and L. pifanoi amastigotes………………. 67 2.2. Evaluation of Leishmania plasma membrane permeabilization by lactoferrin-derived peptides………………………………………………………. 69 2.3. Effect of lactoferrin-derived peptides on L. pifanoi-infected macrophages….. 74 III.2 – Activity of cecropin A-mellitin-derived and lactoferrin-derived peptides against Mycobacterium avium………………………………………………..……….….. 77 1. Material and Methods…………………………………………..…………………………. 79 2. Results………………………………………………………………………………………. 85 2.1. Cecropin A-melittin-derived peptides…………………………………..……….. 85 2.1.1. Effect on the viability of M. avium in axenic cultures…………………….. 85 2.1.2. Effect on M. avium growing inside macrophages and macrophage toxicity……………………………………………………………………….... 85 2.2. Lactoferrin-derived peptides…………………………………………..…………. 87 2.2.1. Effect on the viability of M. avium in axenic cultures…………………….. 87 2.2.2. Effect on M. avium growing inside macrophages and macrophage toxicity……………………………………………………………….………… 88 III.3 – Characterization of peptide-lipid interaction by X-ray diffraction…………… 93 1. Material and Methods…………………………………………………..…………………. 95 2. Results………………………………………………………..…………………………….. 101 2.1. Cecropin A-melittin-derived peptides………………………………………..….. 101 2.1.1. Phase behavior of POPE membranes…………………………………..… 101 2.2. Lactoferrin-derived peptides……………………………………………..………. 106 2.2.1. Phase behavior of POPE membranes…………………………………..… 106 PART IV – DISCUSSION & CONCLUSIONS……………………………………………... 109 Bibliography………………………………………………..………………………………….. 119 3 LIST OF SYMBOLS AND ABBREVIATIONS AIDS – Acquired Immune Deficiency Syndrome AMPs – Antimicrobial Peptides BMMØ – Bone Marrow derived Macrophages CA(1-7)M(2-9) or CAM – hybrid of cecropin A (aa 1-7) and melittin (aa 2-9) CFUs – Colony Forming Units Di-LFampin – Hybrid of two LFampin265-284 coupled by a special lysine linkage Di-LFcin– Hybrid of two LFcin17-30 coupled by a special lysine linkage DMEM – Dulbecco’s Modified Eagle’s Medium FBS – Fetal Bovine Serum FITC – Fluorescein isothiocyanate HBSS – Hank’s Balanced Salt Solution HIFCS – Heat Inactivated Fetal Calf Serum HII – Inverted hexagonal phase HIV – Human Immunodeficiency Virus hLFcin1-11 – lactoferricin peptide containing amino acids 1-11 from human lactoferrin IC50 – Concentration that inhibits 50% of parasites growth at the end of 4h IFN-γ – Interferon gamma IL-10 – Interleukin 10 IL-12 – Interleukin 12 K6 – Lysine Nε-trimethylated CA(1-7)M(2-9) at a lysine in position 6 K7 – Lysine Nε-trimethylated CA(1-7)M(2-9) at a lysine in position 7 LCCM – L929 cell-conditioned medium LD50 – Concentration that inhibits 50% of parasites proliferation at the end of 3 or 5 days for promastigotes and amastigotes, respectively LF – Lactoferrin 4 LFampin265-284 – lactoferrampin peptide containing amino acids 265-284 from bovine lactoferrin LFampin-LFcin – Hybrid of LFampin265-284 and LFcin17-30 LFChimera – Hybrid of LFcin17-30 and LFampin265-284 coupled by a special lysine linkage LFChimera-R – Reverse peptide of LFChimera LFcin17-30 – lactoferricin peptide containing amino acids 17-30 from bovine lactoferrin LFcin-LFampin – Hybrid of LFcin17-30 and LFampin265-284 LPS – Lipopolysaccharide Lα – Fluid lamellar phase Lβ – Gel lamellar phase MAC – Mycobacterium avium complex MTT – 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide P:L – Peptide-to-Lipid molar ratio PBS – Phosphate Buffered Saline PC – Phosphatidylcholine PE – Phosphatidylethanolamine PG – Phosphatidylglicerol POPE – 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine SAXD – Small Angle X-ray Diffraction TB – Tuberculosis TGF-β – Transforming Growth Factor β TH – Lipid transition temperature from fluid lamellar to hexagonal phase TM – Lipid transition temperature from gel lamellar to fluid lamellar phase TNF-α – Tumor Necrosis Factor α WAXD – Wide Angle X-ray Diffraction WHO – World Health Organization 5 ACKNOWLEDGMENTS First of all I would like to thank my supervisors, Margarida Bastos and Salomé Gomes, for all the support, patience, for always being available for helping me and for giving me the opportunity to continue my work. In this year I had the chance not only to grow as a scientist but also as person, and much of that was due to both of them. Also thank to all the people that have helped me during this year making possible the writing of this thesis, especially to:  Guangyue Bai, that although is now very far away her teaching was essential for me.  All the members of the Iron and Innate Immunity group and the Microbiology and Immunology of Infection group from IBMC, especially to Sílvia Costa, Sandro Gomes, António Barroso e Tânia Moniz, your help and friendship was fundamental.  All the people that I have met in Centro Investigaciones Biológicas in Madrid, for making me feel at home even when I was so far away. Especially to the members of the eukaryotic antibiotic peptides group, Luis Rivas for his wisdom, you taught me a lot, to María Fernández-Reyes and María Ángeles Abengózar without whom my work would never be possible and Oscar Santos Calvo, for making me laugh and show me the amazing city that is Madrid.  Daniela Uhríková (Faculty of Pharmacy, Comenius University, Bratislava, Slovak Republic) and Sergio Funari (Desy, Hasylab, Germany) for the introduction to the X-ray world (together with Margarida Bastos) that I completely ignored and now fascinats me.  Jan Bolscher (ACTA, The Netherlands), David Andreu (Universitat Pompeu Fabra, Barcelona) and Paula Gomes (FCUP, Porto) for providing the peptides. Finally, but not least, to my parents and my boyfriend, for their infinite patience and love you are my rocks and the ones that keep my feet on the ground. 12 13 PART I INTRODUCTION 14 15 1. GENERAL INTRODUCTION The introduction of the first antibiotics in the clinical practice in the 1940s (Aminov, R.I. 2009) started one of the greatest successes in the history of medicine. These pharmaceutical compounds saved millions of human lives throughout the last decades and made people believe that all bacterial infections could be treated and they would rapidly become a problem of the past. However this belief was proven to be wrong. After a few years antibiotic-resistant pathogens started to emerge and to disseminate. In response to that, new antibiotics had to be brought to the clinic and, during several decades, the number of deaths caused by infections continued to decline. Nowadays we face a re-emergence of infectious diseases and the appearance of multidrug-resistant “super-bugs”, the treatment of which is increasingly costly and prone to failure. Examples of these multidrug-resistant bacteria are: the methicillin-resistant Staphylococcus aureus (MRSA) that is responsible for a high percentage of hospital-acquired infections and that is spreading outside the hospital zones (Lohner, K. 2009), multidrug-resistant Mycobacterium tuberculosis with approximately 440 000 new cases emerging annually, causing at least 150 000 deaths (WHO 2011a), among many others. This alarming situation was originated from a combination of factors, such as, the excessive and inappropriate use of antibiotics in human and animal health (Lohner, K. & Blondelle, S.E. 2005) and the almost complete absence of discovery of new antibiotics in recent years (only two new classes of antimicrobial compounds were introduced in the market in the last three decades) (Lohner, K. 2009). Figure 1 – Publicity poster from the World Health Organization (WHO) for the World Health Day 2011, where the aim was to reach attention for the antimicrobial resistance problem (WHO 2011c). 16 Considering all this there is an urgent need to develop new and effective antimicrobial therapies. In this context antimicrobial peptides (AMPs) are one potential alternative to fight infectious diseases because their mode of action makes them less prone to resistance induction than currently used antibiotics and permits high activity against a vast range of microorganisms. AMPs are present in almost all living organisms, having broad functions in innate immunity, including immunomodulation, and providing a first line of defense against a wide variety of pathogens (Hancock, R.E. & Sahl, H.G. 2006; Lohner, K. 2009). Their mechanism of action is not fully understood, but it is widely accepted that they act by disrupting the cellular membrane of pathogens, with the possibility of also acting on intracellular targets. Most of these peptides have multiple modes of action, and they work in a multiple-hit strategy which increases their efficiency and capacity to evade potential resistance mechanisms (Nguyen, L.T. et al. 2011). These characteristics may make AMPs useful in the clinic for a longer time, as compared to conventional antibiotics (which become obsolete, due to microbial resistance in 1 to 2 decades on average) (Hancock, R.E. & Sahl, H.G. 2006). 17 2. PATHOGENS In this work the antimicrobial activity of AMPs was assessed on different pathogens, namely Leishmania and Mycobacterium avium. In the next sections a brief introduction will be made to these microorganisms focusing on the epidemiology, structural characteristics, pathology, on the current treatments available and on the status of AMPs as potential agents to fight the diseases inflicted by these pathogens. 2.1. LEISHMANIA Protozoan parasitic diseases, especially malaria, leishmaniasis, and trypanosomiasis, are considered neglected diseases and remain an unsolved public health problem in certain areas of the world, with extremely high death rates. The high mortality in developing countries is often due to the poor sanitary conditions and lack of efficient prophylactic measures, whereas in developed countries, some of these diseases that were eradicated are re-appearing especially as opportunistic diseases in immunocompromised hosts (Piscopo, T.V. & Mallia Azzopardi, C. 2007; Zucca, M. & Savoia, D. 2011). Leishmania is a genus of protozoan parasites that are transmitted to the mammalian host (e.g. canines, rodents and humans) by the bite of phlebotomine sand flies, and are the causative agents of a set of clinical diseases collectively known as leishmaniasis. The World Health Organization estimates that 350 million people are at risk of contracting leishmaniasis and approximately 2 million new cases occur per year. Visceral infections, caused by L. donovani and L. infantum, are highly fatal if not treated, and are responsible for 50 000 deaths annually, a rate surpassed among parasitic diseases only by malaria (WHO 2010). Leishmaniasis is endemic in more than 60 countries worldwide, mostly found in tropical and subtropical regions of the world. This disease is also endemic to all European Mediterranean countries including Portugal, mostly as canine leishmaniasis, in humans the opportunistic co-infection with HIV is of special relevance (WHO 2010). 2.1.1. LEISHMANIASIS Clinical manifestations of leishmaniasis differ widely depending on the Leishmania species (there are at least 20 species) as well as on host factors such as immunity status (Goto, H. & Lindoso, J.L. 2010). The interplay between Leishmania and the mammalian host response is manifest not only in terms of clinical outcome but also in the rate of 18 spontaneous healing and recurrent disease. Three main types of disease patterns occur: cutaneous, mucocutaneous and visceral infections (figure 2) (Piscopo, T.V. & Mallia Azzopardi, C. 2007; Zucca, M. & Savoia, D. 2011). Figure 2 – Clinical forms of leishmaniasis. A) Cutaneous, B) mucocutaneous and C) visceral leishmaniasis (WHO 2011b). Cutaneous leishmaniasis can potentially be associated with all Leishmania species, being the most common form of the disease characterized by the appearance of ulcers at the site of the bite on exposed parts of the body, such as the face, arms and legs. Normally these cutaneous forms of the disease are benign and the lesions spontaneously heal leaving scars. However there are some species of Leishmania that cause more complicated outcomes, for instance mucocutaneous leishmaniasis (WHO 2010). This form of the disease is caused by metastasis to the mucosal tissues of the mouth and upper respiratory tract by lymphatic and haematogenous dissemination potentially evolving to disfiguring lesions. Some Leishmania species that normally do not cause this type of leishmaniasis, have been reported to induce similar lesions in immunosuppressed patients (Goto, H. & Lindoso, J.L. 2010; WHO 2010). Visceral leishmaniasis is the most serious and complicated form of the disease and fatal if not treated. The infection is due to the invasion by the parasites of mononuclear phagocytic cells of the liver, spleen, bone marrow and intestinal mucosa. Aspects such as malnutrition, genetic factors and coinfections that lead to immunosuppression are risk factors to develop the disease (Piscopo, T.V. & Mallia Azzopardi, C. 2007; WHO 2010). 19 2.1.2. LIFE CYCLE The protozoan parasite Leishmania has two major forms: the promastigote, the flagellated form that is transmitted to the vertebrate host by sand fly bites, and the intracellular amastigote, which lives inside mononuclear phagocytic cells of a wide variety of mammals (figure 3). Figure 3 – Life cycle of Leishmania. Schematic representation of the different stages of development of the parasite in the invertebrate host, sand fly, and in the vertebrate host, mammals (Adapted from: Banuls, A.L. et al. 2007; Sacks, D. & Noben-Trauth, N. 2002). In the invertebrate host the promastigote has an elongated shape and a long flagellum. This form infects the mammalian host when a parasitized female sand fly feeds from the blood of the vertebrate. The promastigotes that enter into the mammals organism are opsonized by serum components and then phagocytosed, primarily by macrophages and dendritic cells (Kima, P.E. 2007; Sacks, D. & Noben-Trauth, N. 2002). Inside the phagocytic cells, within the phagolysosome, the promastigotes undergo several structural and morphological changes to give rise to the intracellular amastigotes, which are ovoid and nonflagelated. Amastigotes proliferate within the phagolysosomes (or parasitophorous vacuoles) where they reproduce by binary fission until the number of 20 parasites increases enough for the cell to burst and free the amastigotes to infect other cells and continue their life cycles. The parasites could then go back to the invertebrate host when a sand fly ingests blood that contains mononuclear phagocytic cells infected with amastigotes (Banuls, A.L. et al. 2007; Kamhawi, S. 2006; Olivier, M. et al. 2005). One of the most remarkable accomplishments of Leishmania is that they successfully live inside mammalian cells that are responsible for killing pathogens, the macrophages. For that Leishmania had to evolve a range of sophisticated mechanisms to subvert host surveillance by altering the macrophage signal transduction machinery, thereby modulating the macrophage environment in its favor. Some examples of these mechanism are the production of lipophosphoglycan at the surface of the parasites that confers resistance to the lytic action of the complement system, allowing the infection of macrophages and creating appropriate conditions for the promastigote-to-amastigote differentiation (Lodge, R. & Descoteaux, A. 2005). Other examples include preventing the activation of macrophages defense mechanisms such as the production of oxygen and nitrogen reactive species and also inhibition of an adequate and protective T cell response by inhibiting the production of pro-inflammatory signals such as IL-12 and promoting the release of anti-inflammatory cytokines, like IL-10 and TGF-β. (Kima, P.E. 2007; Olivier, M. et al. 2005). 2.1.3. TREATMENTS Drug treatment for leishmaniasis exists since the 20th Century, but only a limited range of drugs are available. The efficacy of these drugs is affected by a combination of factors, such as the differences in the sensitivity of Leishmania species to the drugs and their pharmacokinetic properties, and the immune status of the patient (Goto, H. & Lindoso, J.L. 2010). The current treatment of leishmaniasis consists in the administration of pentavalent antimonials as a first option. As second line drugs there are some alternatives such as amphotericin B, which is effective in antimonials-resistant cases, miltefosine which was the first orally active drug made available against leishmaniasis, pentamidine that can also be used in the treatment of resistant cases of visceral leishmaniasis, as well as paromomycin (Goto, H. & Lindoso, J.L. 2010; Piscopo, T.V. & Mallia Azzopardi, C. 2007; WHO 2010). Parasite resistance to these conventional treatments, lack of efficacy, toxicity and high costs associated with them lead to a clear need to improve the existing drugs and to 21 identify new targets and in that way develop new therapies (Piscopo, T.V. & Mallia Azzopardi, C. 2007). However, the highest prevalence of parasitic diseases such as leishmaniasis occurs in the poorest areas of the world, and from the total investment in health research in the world, the work related with malaria, leishmaniasis and trypanosomiasis accounts for only about 0.1% whereas the contribution of these diseases to total burden is 5%. Nevertheless in the last years the search for new antiparasitic treatments has received new impulse thanks to new technical and political developments, and the appearance of new therapies, such as protease and topoisomerase inhibitors, RNA interferencebased approaches, nano-drug delivery and proteomics, as well as antimicrobial peptides (Zucca, M. & Savoia, D. 2011). 2.1.4. LEISHMANIA AS A TARGET OF AMPS In what concerns AMPs, protozoans have received much less attention as a target than pathogens such as bacteria or fungi, although a variety of in vivo and in vitro antiparasitic assays suggest that these compounds could represent a powerful tool for the development of new therapies as well as to complement current ones (Rivas, L. et al. 2009). Following the discovery of AMPs, magainins and cecropins were the first antimicrobial peptides to display antiparasitic activities along with hybrids of cecropin A-melittin (Andreu, D. et al. 1992; Boman, H.G. et al. 1989; Mor, A. 2009). Histatin 5, a human salivary AMP, is also active against Leishmania, not by acting in the plasma membrane, as usual for AMPs, but by decreasing the mitochondrial ATP synthesis leading to cell death (LuqueOrtega, J.R. et al. 2008). Although Leishmania are eukaryotic pathogens, their cytoplasmic membranes have higher levels of negative phospholipids, such as phosphatidylserine and phosphatidylinositol than higher eukaryotes, conferring them more susceptibility to the action of AMPs than mammal cells (Wassef, M.K. et al. 1985). Leishmania is an interesting model as a target for AMPs primarily due to the absence of barriers external to the plasma membrane contrary to what happens in bacteria. The promastigote form of the parasite is endowed with an anionic glycocalix, essentially composed of lipophosphoglycan (LPG), an anionic oligosaccharide, anchored to the membrane through glycosylinositolphospholipids (GIPLs) (figure 4). Abundant proteolytic activity also exists at the surface, due to the presence and activity of GPI-proteins and proteophosphoglycans (PPGs) (figure 4) (Rivas, L. et al. 2009). These structures present at the surface of the membrane have some duality in what concerns resistance and/ or susceptibility to AMPs. The presence of a negatively 28 29 3. ANTIMICROBIAL PEPTIDES (AMPS) Considering the potential of antimicrobial peptides as new alternative therapies to fight infectious diseases, in the next sections, several aspects concerning AMPs will be addressed, namely: their structural characteristics, mechanisms of action, role in the immunity, and the problems, successes and challenges of introducing AMPs in the clinical practice. The term “Antimicrobial peptide” refers to a large number of peptides first characterized on the basis of their activity against bacteria and fungi. They constitute a primitive immune defense mechanism and they were found, first in the 1980s, on a variety of eukaryotic organisms, from insects to humans (Nguyen, L.T. et al. 2011; Reddy, K.V. et al. 2004). Some of them were first isolated due to their antimicrobial activity, whereas others were discovered for various unrelated functions before their potential as antimicrobial peptides was recognized (Wiesner, J. & Vilcinskas, A. 2010). These peptides are multifunctional and act in concert with other immune mechanisms and have evolved in nature throughout centuries to protect their hosts against diverse pathogens, such as bacteria, fungi, virus, protozoa and even cancer cells (Nguyen, L.T. et al. 2011). So far, more than 1500 peptides from different sources have been reported in The Antimicrobial Peptide Database – APD (http://aps.unmc.edu/AP/main.php) and the majority of them are classified as antibacterial peptides. Virtually all human tissues can produce AMPs, either constitutively or induced by inflammatory stimulus, more frequently in body sites exposed to pathogens, such as the skin or mucosa, or in some blood cell types, such as neutrophils, eosinophils and platelets (Wiesner, J. & Vilcinskas, A. 2010). Natural AMPs are gene-encoded and ribosome-synthesized peptides that derive from precursor peptides through one or more proteolytic activation steps. De novo design of synthetic peptides was introduced in the 1990s with the use of high-throughput combinatorial library screening, structure based modeling and predictive algorithms to determine the optimal active peptide. Although this approach produced interesting results, the need for a high initial investment in technology compared to traditional methods has delayed the widespread use of this approach. On the other hand, rational design of peptides using existent sequences as templates that are modified to obtain peptides with the desired properties (antimicrobial activity and selectivity) and the discovery of AMPs through the search within natural proteins (e.g. lactoferrin) have also led to the successful production of more potent antimicrobial peptides (Nguyen, L.T. et al. 2011). 30 3.1. STRUCTURAL CHARACTERISTICS Antimicrobial peptides vary widely in many aspects, such as length, sequence, structure and source, but there are some common traits for most of them, that are important for their antimicrobial activity. They can be defined as short (<50 amino-acids), with a positive net charge (generally +2 to +9, due to the existence of lysine and arginine residues) that allows them to interact with negatively charged membranes, such as bacterial membranes, although anionic antimicrobial peptides have also been described. They also have a substantial proportion of hydrophobic residues. AMPs have a random structure in solution but when in a membranous environment they adopt an amphipatic structure, allowing them to bind more efficiently to phospholipid membranes (Shai, Y. et al. 2006). The full classification of AMPs is difficult due to their huge diversity but almost all authors divide and classify the peptides based on their secondary structure. One of those classifications divides AMPs into three groups (table 1 and figure 7): α-helical peptides, β-sheet peptides and extended peptides which do not fold into regular secondary structures and are often composed of a high number of certain amino-acids, like histidine, tryptophan and arginine (Nguyen, L.T. et al. 2011). Table 1 – Overall characteristics of some AMPs (adapted from: Van 'T Hof, W. et al. 2001). Group Peptides Organism Source Antimicrobial activity a Reference α-helical peptides melittin bee venom (apitoxin) G-; G+; V; H (Reddy, K.V. et al. 2004) magainins frogs skin mucus G-; G+; F (Nguyen, L.T. et al. 2011) cecropins insects: moth hemocytes G-; G+; F; P (Reddy, K.V. et al. 2004) cathelicidins LL-37 humans leukocytes, epithelia G-; G+ (Guaní-Guerra, E. et al. 2010; Nguyen, L.T. et al. 2011) Dermaseptins frogs skin mucus G-; G+; F (Van 'T Hof, W. et al. 2001) β-sheet peptides lactoferricin cow, humans digested lactoferrin G-; G+; V; F; P; T (Chan, D.I. et al. 2006) defensins humans leukocytes, epithelia G-; G+; V; F (Guaní-Guerra, E. et al. 2010) protegrins pig leukocytes, epithelia G-; G+; F (Reddy, K.V. et al. 2004) extended peptides indolicin cow leukocytes G-; G+; V; F; P (Chan, D.I. et al. 2006) tripticin pig leukocytes G-; G+; F (Chan, D.I. et al. 2006) histatins humans saliva G-; G+; F (Van 'T Hof, W. et al. 2001) PR-39 pig intestine G-; G+; (Chan, D.I. et al. 2006) a G-: Gram-negative bacteria; G+: Gram-positive bacteria; F: fungi; P: protozoan; T: tumor; H: hemolytic. 31 Figure 7 – The three major structures of AMPs. A) α-helical peptides B) β-sheet peptides C) extended peptides. Positively charged side chains are colored in blue, negatively charged side chains in red and remaining side chains in grey. PDB IDs: magainin 2: 2MAG; bovine lactoferricin: 1LFC; indolicidin: 1G89 (adapted from: Nguyen, L.T. et al. 2011). 3.2. MECHANISMS OF ACTION The mechanism of action of antimicrobial peptides is not yet fully understood but it is well known that the biophysical properties mentioned above (secondary structure, overall charge and hydrophobicity) play important roles in their interaction with pathogen membranes. Other parameters are important for their action, such as the membrane lipid composition, the peptide-to-lipid molar ratio, and environmental conditions like ionic strength and pH (Lohner, K. & Blondelle, S.E. 2005). Due to the high diversity of AMPs and their properties there is no universal mechanism for the action of all AMPs as they can act in multiple ways making difficult to unravel all the molecular events resulting from the interaction between peptides and membranes. The activity of AMPs is most probably not related to specific membrane protein receptors since studies using all D-amino acids instead of L-amino acids in magainin, cecropin and melittin, showed that these D-peptides exhibited similar activities than the natural ones, suggesting that the interaction between peptides and membranes does not depend on chiral centers (Merrifield, E.L. et al. 1995; Wade, D. et al. 1990). There are some proposed mechanisms, providing explanations of possible ways for the peptides to disrupt the membrane leading to cell death. Overall, all the proposed mechanisms rely on the same main factors for initial action: adsorption of AMPs onto the membrane due to electrostatic interactions between the cationic peptides and the headgroups of anionic phospholipids, and then they vary in the next steps of action. According to the barrel stave model (figure 8), peptides large enough (~ 22 amino acids) form transmembrane aqueous channels/pores by inserting into the hydrophobic core perpendicularly to the membrane plane where the hydrophobic region of the peptide is aligned with the acyl chains of the phospholipids while the peptide hydrophilic regions form the inner surface of the pore channel (figure 8). The continuous recruitment of 32 peptides to the membrane increases the pore size leading to leakage of the cells contents, membrane depolarization and thereby cell death (Nguyen, L.T. et al. 2011). The peptides arrangement in the membrane causes high repulsion due to their charges and this can culminate in the disintegration of the pore. This model is only possible for peptides with not too high charge and it is believed that the number of peptides acting by this mechanism is very low. An example is alamethicin from the fungus Trichoderma viridis, and perforin, produced by mammalian killer lymphocytes and complement component C9 (Wiesner, J. & Vilcinskas, A. 2010). The carpet model (figure 8) proposes that the membrane surface is covered by the peptides, aligned parallel to the bilayer surface, in a “carpet like” manner, diminishing the fluidity of the membrane. When the concentration of the peptide reaches a threshold value the integrity of the membrane is lost leading to membrane micellization, toroidal pores or aggregated channels formation (see below) and cell death (figure 8) (Lohner, K. 2009). The formation of toroidal or wormhole pores (figure 8) is the best characterized and the most consistent mechanism of action of AMPs. This model can be seen as representing one of the possible final stages of the previous carpet model, where the formation of pores leads to disruption of the membranes. In fact, these two models have been unified in the so called “Shai-Matzusaki-Huang” model that proposes that peptides bind parallel to the membrane with the apolar amino acids penetrating partly into the hydrophobic core, while the cationic residues interact with the headgroups of anionic phospholipids, inducing membrane thinning and a curvature strain. To release this strain the orientation of the peptides changes from parallel to perpendicular, inducing bending of the membrane interface towards the hydrophobic interior, maintaining the contact between the peptides and the charged headgroups from the phospholipids. These events lead to the formation of toroidal pores composed by the peptides and the lipid headgroups (figure 8). Upon disintegration of the pores some peptides can be translocated to the inner leaflet of the membranes reaching the cytoplasm or the membrane can be disrupted due to depolarization or micellization, resulting in cell death (Nicolas, P. 2009; Rivas, L. et al. 2009). Other mechanisms of action have recently been suggested that do not imply the recruitment of a high concentration of peptides to the membrane, and they do not necessarily exclude each other (or even the models explained above) (figure 8). Some examples are, lipid segregation, where the lipids can move laterally in the membrane and thus form domains rich in anionic lipids (induced by the presence of the positively charged peptide), and this can induce small leakage of intracellular contents and/ or 33 membrane destabilization and depolarization due to dissipation of the membrane potential; formation of non-bilayer intermediates in the membrane and non-lamellar structures; and membrane permeabilization and/ or depolarization due to an efflux of small anions across the bilayer by coupling with peptides. (Nguyen, L.T. et al. 2011). Whatever the mechanism is, peptide interaction with membranes can culminate in cell death by the dissipation of transmembrane electrochemical ion gradients, loss of metabolites, and eventually lysis of the cell (Wiesner, J. & Vilcinskas, A. 2010). If the peptides are translocated to the cytoplasm, with or without permeabilising the cytoplasmic membrane, some other effects could arise due to the existence of intracellular targets, such as DNA, RNA or protein synthesis and folding. Buforin is an example of a non-lytic peptide which although it has a high antimicrobial activity this does not correlate with lysis of the bacterial membrane. Instead the peptide translocates efficiently to the cytoplasm probably through the formation of short-lived toroidal pores, enabling this peptide to enter the cell and act on intracellular targets (Henriques, S.T. et al. 2006; Kobayashi, S. et al. 2004; Kobayashi, S. et al. 2000). Figure 8 – Proposed mechanisms of action for AMPs. Illustration of some of the proposed events/mechanisms occurring at the cytoplasmic membrane after a initial interaction of antimicrobial peptides with the membranes (Nguyen, L.T. et al. 2011). 34 AMPs can also be cytotoxic against cancerous cells promoting apoptosis (e.g. lactoferrin proved to inhibit the tumor growth and metastasis in a rodent model with breast cancer, while lactoferrricin causes apoptosis in Jurkat T-leukemia cells) (Iigo, M. et al. 2009; Mader, J.S. et al. 2007; Wiesner, J. & Vilcinskas, A. 2010). It is believed that this effect is first due to lysis of the cellular membrane of the cancerous cells, that are recognized by some AMPs due to the unusually high amounts of negatively charged gangliosides and phosphatidylserine (PS) in the surface of these cells, a negative membrane potential and a higher membrane fluidity, allowing the peptides to insert into the membrane more easily (Hoskin, D.W. & Ramamoorthy, A. 2008). Afterwards, mitochondrial membranes can also be disrupted due to entrance of the peptide into the cell, apoptosis pathways can be activated and angiogenesis inhibited (Gifford, J.L. et al. 2005; Guaní-Guerra, E. et al. 2010). 3.3. SELECTIVITY / TOXICITY Independently of how AMPs exert their activity against pathogens the peptides have to interact with the cellular membrane resulting either in disruption or traversing this barrier. This interaction must be as selective as possible regarding the distinction between mammalian cells (eukaryotic cells) and pathogen cells, such as bacteria (prokaryotic cells). Cytoplasmic membranes of mammalian cells are predominantly constituted by phospholipids and cholesterol. Phospholipids are asymmetrically distributed between the inner and the outer leaflet of the bilayer exposing predominantly zwitterionic phosphatidylcholine (PC) and sphingomyelin to the extracellular side. On the other hand, cytoplasmic bacterial membranes do not have cholesterol and are mainly composed of zwitterionic phosphotidylethanolamine (PE) and negatively charged phosphatidylglicerol (PG), diphosphatidylglicerol and cardiolipin, conferring an overall negative charge to the outer side of the cytoplasmatic membrane (Lohner, K. 2009; Lohner, K. & Blondelle, S.E. 2005). The main principle of peptide interaction with membranes relies on electrostatic interactions between cationic peptides and negatively charged membranes and thus considering the composition of both cell types, bacteria are a preferential target for AMPs. However, before reaching the cytoplasmic membrane of the bacteria, peptides have to overcome other barriers. In the case of Gram-positive bacteria, a thick peptidoglycan layer embedded with teichoic and lipoteichoic acids (which are negatively charged) is present around the cell, exterior to the lipid membrane (figure 9). Gram-negative bacteria also have a complex cell wall consisting of an outer lipid membrane layer, with a unique and highly asymmetrical composition, and a layer of peptidoglycan (much thinner than that of Gram-positive bacteria) between the inner and the outer membranes (figure 9) (Cabeen, 35 M.T. & Jacobs-Wagner, C. 2005). In Gram-negatives outer membrane, lipopolysaccharides (LPS – negatively charged) are located exclusively in the outer leaflet and phospholipids are confined to the inner leaflet (figure 9) (Lohner, K. 2009; Lohner, K. & Blondelle, S.E. 2005). In the case of Mycobacteria, an even thicker and more complex cell wall is also present (see section 2.2.1). Even so, in normal conditions the overall charge of a bacterial cell wall is negative, increasing the chances of interaction with antimicrobial peptides. (Guenin-Mace, L. et al. 2009; Prescott, L.M. et al. 2002). Figure 9 – Bacteria cell wall. Schematic representation of the cell wall of a) Gram-positive and b) Gram-negative bacteria (Cabeen, M.T. & Jacobs-Wagner, C. 2005). The membrane potential can also contribute for the selectivity of AMPs since in eukaryotes it is less negative than in prokaryotes (Van 'T Hof, W. et al. 2001). Antimicrobial peptides can also kill eukaryotic pathogens, such as fungi and protozoa, raising the question of what are the features of the pathogens, either eukaryotes or prokaryotes that allows some AMPs to discriminate them from mammal cells. These specificity seems to rely on charge, since pathogens seems to have a higher percentage of anionic phospholipids than mammal cells, thus explaining the specificity of the cationic AMPs (Lohner, K. 2009; Lohner, K. & Blondelle, S.E. 2005). However some peptides do not distinguish host and pathogen cells being able to disrupt mammalian cells (e.g. melittin) and thus cannot be used in the clinic. The reasons behind this activity are not completely understood and are most probably due to the properties of the peptides, like a 36 high positive charge, too large hydrophobic surfaces, between others (Nguyen, L.T. et al. 2011). In the evaluation of the activity of AMPs the cytotoxicity of these peptides towards mammalian cells should always be addressed. There are many different cytotoxicity assays that can be performed but the outcome of each of them strongly depends on many factors such as the origin and life storage of the mammal cells, the peptide-to-cell ratio, the medium used, between others, so the results must be carefully analyzed (Van 'T Hof, W. et al. 2001). Based on the results of cytotoxicity assays together with the antimicrobial activity of AMPs, the selectivity index, defined as the cytotoxicity activity divided by the antimicrobial activity, can be determined being a useful tool to predict the potential of a given AMP as a therapeutic agent. However, the characteristics of these peptides and the prediction of their action, either antimicrobial activity or cytotoxicity, is not straightforward and so they should be analyzed on a case-by-case basis. 3.4. AMPS IN THE IMMUNITY Besides the role of AMPs as endogenous antibiotics (directly killing the pathogens), they can also participate in multiple aspects of immunity and for that they can also be called host defense peptides (HDPs). Combination of these different but complementary functions is essential for the effective control of infections in the host organism. AMPs seem to actively participate in the immune system either by immunomodulation, neutralization of endotoxins, enhancement of phagocytosis, induction of both angiogenesis and wound repair, leukocyte chemotaxis and synergism with cytokines (Yang, D. et al. 2002; Zaiou, M. 2007). Thus, in response to an infection, antimicrobial peptides can promote bacterial clearance not only through direct killing but also through the establishment of immune cell circuits (Auvynet, C. & Rosenstein, Y. 2009; GuaníGuerra, E. et al. 2010). As a consequence of the actions stated above these peptides are also involved in autoimmunity. The abnormal concentration, processing or signaling of AMPs is associated with a growing list of autoimmune diseases (figure 10), in which these peptides usually act, not as a single direct cause but as an important factor that influences the outcome of the diseases (Guaní-Guerra, E. et al. 2010). 37 Figure 10 – Role of AMPs in the immune system. Exemplification of some of the immune processes and inflammatory diseases where AMPs are involved (Guaní-Guerra, E. et al. 2010). 3.5. RESISTANCE TO AMPS Induction of resistance is less likely for AMPs when compared to conventional antibiotics, due to the fast kinetics of the antimicrobial process and to the nature of the target – plasma membrane – since it would require an overall reorganization of the cell membrane structure, namely the phospholipid composition, affecting simultaneously the pleiade of transport systems and the enzymes embedded in the phospholipid matrix. Although in principle acquiring resistance to AMPs is more difficult, it has been pointed that it might have more severe effects if it led to cross-resistance to innate human antimicrobial peptides (Hancock, R.E. & Sahl, H.G. 2006). The most common mechanisms of bacterial resistance to AMPs described so far consist in the modification of the bacterial envelope leading to charge reduction and the proteolytic cleavage of peptides (Wiesner, J. & Vilcinskas, A. 2010). These mechanisms include the production of secreted proteins or cell-surface proteins that irreversibly binds or cleaves AMPs and glycopolymeric matrices that trap the peptides preventing their access to the bacterial cytoplasmic membrane (figure 11). Electrostatic repulsion of AMPs can also arise from several modifications such as substitutions of lipid A of LPS by aminoarabinose and ethanolamine, modification of phosphatidylglicerol by ligation of a L- 44 Lactoferricin is obtained by pepsin digestion of lactoferrin, and has been extensively studied. The bovine peptide is constituted by 25 amino acids, corresponding to the amino acids 17-41 in the native protein (bovine lactoferrin) it has a disulfide bond between two cysteine residues, and forms a β-sheet structure (Chan, D.I. et al. 2006; Gifford, J.L. et al. 2005). Antimicrobial activity of lactoferricin ranges from several Gram-negative to Grampositive bacteria, (Jenssen, H. & Hancock, R.E. 2009; Oo, T.Z. et al. 2010; SanchezGomez, S. et al. 2011) fungi, protozoan and virus. Lactoferricin has been shown to have antitumor effects, like inhibition of tumor metastasis, suppression of tumor-induced angiogenesis and significant reduction of solid tumor (fibrosarcomas, melanomas, coloncarcinomas) size in mice, without affecting erythrocytes or fibroblasts (Eliassen, L.T. et al. 2002; Gifford, J.L. et al. 2005; Yoo, Y.C. et al. 1997a; Yoo, Y.C. et al. 1997b). This peptide can also have immunomodulatory properties, playing a role in the innate and adaptive immune system, and it was also proposed that it inhibits septic shock by binding to endotoxins (Chan, D.I. et al. 2006; Yamauchi, K. et al. 1993). J. Groenink, et al in 1999 synthesized shorter peptides with sequences homologous to the N-terminal domain of bovine and human lactoferricin, and found that of the peptides tested, bovine lactoferricin containing amino acids 17-30 (LFcin17-30) had the higher number of positively charged residues and the highest antimicrobial activity against both Gram-positive and Gramnegative bacteria (Groenink, J. et al. 1999). The corresponding human peptide, hLFcin with the first 11 amino acids from human lactoferrin (hLFcin1-11) was demonstrated to be active against a variety of bacteria, such as Acinetobacter baumanii and Stahphylococcus aureus and fungi, especially Candida albicans. A clinical phase I trial using this peptide to combat bacterial hospital infections has been completed, and the overall safety was proved in preclinical and clinical studies (table 2) (Wiesner, J. & Vilcinskas, A. 2010) (www.am-pharma.com). Recently, another peptide in the N1 domain of bovine lactoferrin was identified by the group of Jan G. Bolscher, namely lactoferrampin containing the amino acids 268-284 (LFampin268-284). This peptide exhibited antimicrobial activity against a broad range of pathogens, different from the activity of lactoferricin, especially against Candida albicans and several bacteria, (Van Der Kraan, M.I. et al. 2004). Systematic studies, using the initial sequence of lactoferrampin to obtain other peptides by truncation and/ or extension, have shown that lactoferrampin 265-284 (LFampin265-284) was the shortest and most active peptide, especially against C. albicans (Van Der Kraan, M.I. et al. 2005a; Van Der Kraan, M.I. et al. 2005b). The three extra amino acids (Aspartic acid-Leucine-Isoleucine) did not confer any additional positive charge (one of the most important characteristics for AMPs) since the first is negatively charged and the other two are uncharged. Instead, this 45 N-terminal sequence endowed lactoferrampin with a high tendency to adopt a more stable amphipatic α-helix, enhancing in this way its microbicidal activity (Haney, E.F. et al. 2009; Van Der Kraan, M.I. et al. 2005a; Van Der Kraan, M.I. et al. 2006). Lactoferricin and lactoferrampin are spatially close in lactoferrin, making it plausible that they can cooperate in many of the beneficial properties of this protein. To test if these peptides would form a functional unit, a chimeric peptide (LFChimera) containing LFcin17-30 and LFampin265-284 was synthesized by Jan G. Bolscher (Bolscher, J.G.M. et al. 2009). To try to mimic the spatial topology of these two peptides in lactoferrin, LFcin17-30 and LFampin265-284 are coupled by their C-terminals to the αand ε-amino groups, respectively, of an additional lysine leaving the two N-terminals as free ends (figure 15). LFChimera displays a strong activity against a wide variety of pathogens such as Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus, Vibrio parahaemolyticus, Entamoeba histolytica, which is maintained in different physiological conditions (including high ionic strength and rich growth medium) (Bolscher, J.G.M. et al. 2009; Flores-Villasenor, H. et al. 2010; Leon-Sicairos, N. et al. 2009; Lopez-Soto, F. et al. 2010). Figure 15 – Lactoferrin-derived peptides. A) Ribbon diagram of bovine lactoferrin with LFcin1730 (yellow) and LFampin265-284 (red). B) Designed of LFChimera composed by LFcin17-30 and LFampin265-284 coupled by an additional lysine (K) (adapted from: Bolscher, J.G.M. et al. 2009). 46 47 4. LIPIDS IN BIOLOGICAL AND MODEL MEMBRANES Biological membranes are mainly composed by a lipid matrix that surrounds all other components, like proteins, glycoproteins and glycolipids that performs crucial functions to the life of the cell. Phospholipids, the building blocks of the cellular compartments, are amphiphilic molecules composed by a hydrophilic headgroup and a hydrophobic tail that consist in the hydrocarbon chains. The lipid composition is different between organisms, and even cell types of the same organism. The phospholipid distribution between the inner and outer leaflets of the lipid bilayer is asymmetrical, and some lipids are richer in one leaflet than the other (Pozo Navas, B. et al. 2005). Although, in the past, the widespread opinion was that lipids are passive components in biological membranes and that only proteins have specific physiological functions, it is now known that lipid molecules are dynamic and actively involved in various biological processes being not only a scaffold for proteins (Haney, E.F. et al. 2010; Luzzati, V. 1997). 4.1. LIPID POLYMORPHISM The way that lipids are structurally organized when in contact with water, is highly variable, depending on their intrinsic properties, on the composition of the medium (pH, ionic strength, additives, etc), as well as on temperature and water content, giving rise to a complex phase-diagram (figure 16). Figure 16 – Temperature-composition phase diagram of monoolein. The phases formed by the lipid monoolein depend on temperature and hydration as illustrated in this phase diagram (adapted from: Qiu, H. & Caffrey, M. 2000). 48 This high polymorphism allows them to self-assemble into different phases and threedimensional ordered structures, that are usually classified as lamellar, and non-lamellar (such as hexagonal and cubic) phases. These non-lamellar phases were first characterized by Luzzati and coworkers using X-ray scattering techniques to analyze the three-dimensional arrangement of numerous membranes (Haney, E.F. et al. 2010). Different lipids have different propensity to form the different lipid phases mainly due to the relative size of their polar headgroup and apolar tails, that are responsible for determining the size and shape of the lipid molecule (figure 17) (Tresset, G. 2009). In cases where the headgroup and the hydrocarbon chains have similar cross-sectional areas, the molecule has a cylindrical shape (e.g. phosphatidylcholine - PC and phosphatidylserine). Lipids with a small headgroup like phosphatidylethanolamine (PE) have the structure of a truncated cone. In contrast when the hydrophobic part occupies a relatively smaller surface area, the molecule has the shape of an inverted cone (e.g. lysophosphatidylcholine) (figure 17) (Sprong, H. et al. 2001). Figure 17 – Phospholipids molecular shapes: inverted cone (LPC - lysophosphatidylcholine), cylindrical (PC - phosphatidylcholine) and truncated cone (PE - phosphatidylethanolamine), and their role in the shape of the cellular membrane (adapted from: Sprong, H. et al. 2001). The functional structure of a biological membrane is a lamellar lipid phase that acts as a fundamental permeability barrier. Lamellar phases are characterized by two opposing lipid monolayers with the hydrocarbon chains facing each other. Two types of lamellar phases can occur, a liquid crystalline or fluid lamellar phase (Lα) that occurs at higher temperatures, where the hydrocarbon chains are in a melted and fluid state allowing the membrane to spontaneously reorganize upon external stimulus, and a gel lamellar phase (Lβ) (lower temperatures) in which the hydrocarbon chains are extended and rigid (figure 18) (Haney, E.F. et al. 2010). The typical state in biological membranes is the liquid crystalline or fluid lamellar state. Lamellar phases typically occur for lipids with a packing parameter (p) close to 1, which means that an individual molecule fits to a cylinder (e.g. PC) (figure 17) (Tresset, G. 2009). 49 Figure 18 – Lamellar phases. Schematic representation of the structure and orientation of the phospholipids hydrocarbon chains in the two lamellar phases, gel (Lβ) and fluid (Lα) (adapted from: Tresset, G. 2009). Although the maintenance of a stable bilayer is essential for normal membrane function, it is well known that in some cases, membranes containing high amounts of non-lamellar phase-forming lipids (e.g. phosphatidylethanolamine) have the ability to form more complex 3D morphologies, non-lamellar structures such as hexagonal and cubic phases, that are believed to play important roles in biological processes (Lohner, K. 2009; Lohner, K. & Blondelle, S.E. 2005). Hexagonal phases are made of lipids that all together adopt a cylindrical shape. When the phospholipids headgroups are oriented towards the core of the cylinder, that is filled with water, the structure is called inverted hexagonal phase – HII, whereas in a normal/ micellar hexagonal phase – HI, the centers of the cylinders consist of the hydrocarbon chains or phospholipid tails (figure 19). The HII is the simplest non-lamellar structure implied in membrane fusion processes. Very few lipids adopt an HI phase (Haney, E.F. et al. 2010; Tresset, G. 2009). The transition from lamellar to hexagonal phase occurs with increasing temperature as the interfacial area expands leading to unfavorable contact between aqueous and hydrophobic regions resulting in the formation of inverted hexagonal phases (Rappolt, M. et al. 2003). Figure 19 – Hexagonal phases. Adopted orientation of phospholipids in the different hexagonal phases HI and HII. In red the headgroups of phospholipids (hydrophilic region) and in grey the hydrocarbon chains (hydrophobic region) (Tresset, G. 2009). 50 Other important non-lamellar structures are the three-dimensional cubic phases. The cubic symmetry is attributed not only to a lipid phase but to a large family of phases that are complex and diverse (Luzzati, V. et al. 1997). Several cubic phases are known and they can be divided essentially into two classes, bicontinuous and micellar phases. The bicontinuous phases are unique and consist on a single bilayer folded into threedimensional cubic network separating two disjointed water compartments with continuous regions of both polar (hydrophilic headgroups) and non-polar (hydrocarbon chains) structures (Ia3d, Pn3m, Im3m) whereas the micellar phases are impervious to watersoluble components and the structure is made of disjointed micelles with different sizes allowing a more efficient packing on a cubic lattice (e.g. Fd3m, Pm3n) (figure 20) (Luzzati, V. 1997; Tresset, G. 2009). Figure 20 – Cubic phases of lipids. Bicontinuous cubic phases: Im3m, Pn3m and Ia3d; and a micellar cubic phase (Fd3m) (adapted from: Tresset, G. 2009). Growing evidences show that cubic phases are ubiquitous in the biological world as they have been detected in the plasma membrane of archaebacteria, as well as in the endoplasmic reticulum and mitochondria of mammalian cells. These phases are also involved in biological processes such as membrane fusion, fat digestion and in the reorganization of cell membrane composition (Almsherqi, Z.A. et al. 2009; Erbes, J. et al. 1994; Tresset, G. 2009). 51 4.2. AMP INTERACTION WITH LIPID MEMBRANES AS REVEALED BY MODEL STUDIES Regarding the interaction of antimicrobial peptides with membranes it is crucial to understand if AMPs are capable of altering the phospholipid phase behavior and in that case to correlate these possible changes in lipid polymorphism with models for the mechanism of action of AMPs. Using model membranes, namely liposomes (concentric bilayer vesicles) with different phospholipid compositions, and analyzing them by using different techniques (e.g. calorimetry, X-ray diffraction, spectroscopy), can give the desired information that enables the understanding of these changes and how they affect the overall structure and stability of the membrane. In this way the mechanism of action of AMPs could be clarified, and lead to a fine-tuning of new peptides in order to obtain a better therapeutic agent. The induction of cubic phases by AMPs, mostly by lowering the lamellar to non-lamellar phase boundary of lipid membranes, could mean that the peptide induces membrane disruption. Without disruption of the membrane, the formation of nonlamellar phases can also alter the lipid domains in the cellular membrane that are important for many biological processes such endocytosis (induced by lipid rafts) or even the activity of membranar proteins that most of the time depend on these domains culminating into impairment of the membrane function (Lohner, K. 2009; Lohner, K. & Blondelle, S.E. 2005). Model membranes can be composed of different phospholipids, and as such their characteristics can be fine-tunned to the desired purpose of study. The compositions most widely used for the study of AMPs are: zwitterionic membranes of PC that mimics the erythrocytes membranes; the combination of PC and PG or PE with PG to try to mimic the cytoplasmic bacterial membrane that has a global negative charge. Recently much emphasis has been put on the use of POPE as the major class of phospholipids in the bacterial membrane whereas the amount of PC is more important in fungus. These are very simplified models of more complex structures, where the main aspect relies on changing the charge balance. Nevertheless, it is more and more clear that the AMPs are sensitive to the actual membrane composition, and thus efforts are also being made in biophysical studies to vary model membrane composition. More elaborated models can be used by the introduction of sphingomyelin and sterols for eukaryotic membranes and cardiolipin for prokaryotic membranes, or even by using membrane extracts of the bacteria. It should nevertheless be stressed that the main paradigm of model studies is to “keep it simple”, in order to enable discriminate analysis of interaction effects. 52 53 PART II OBJECTIVES 60 61 1. MATERIAL AND METHODS Peptides The lactoferrin-derived peptides tested for their antimicrobial activity against Leishmania were lactoferricin (LFcin17-30), lactoferrampin (LFampin265-284), LFChimera, and LFChimera surrogates described on table 4 of page 56. Leishmania donovani promastigotes Promastigotes of strain MHOM/SD/00/1S-2D (WT TURCO) of L. donovani were grown at 26 ºC in RPMI 1640 (Gibco) pH=6.8, supplemented with 10% heat-inactivated fetal calf serum (HIFCS), 25 mM HEPES, 2 mM L-glutamine, 10 μg/mL gentamicin and 20 units/mL unicilin. Promastigotes of the WT TURCO-derived 3-Luc strain of L. donovani, which express a cytoplasmic form of luciferase (Luque-Ortega, J.R. et al. 2001), were grown in the same conditions as the wild-type strain, except for the addition of 30 μg/mL geneticin to the medium. Leishmania pifanoi amastigotes Amastigotes MHOM/VE/60/Ltrod of L. pifanoi were grown at 32 ºC in M199 pH=7.2, supplemented with 20% HIFCS, 5% trypticase, 13.8 mM glucose, 76 μM hemin and 48 μg/mL gentamicin. Leishmania harvesting The parasites were collected at late exponential growth phase by centrifugation (2000 rpm, 10 min, 4 ºC). The supernatant was discarded carefully and the cells were washed twice with Hank’s Balanced Salt Solution (HBSS: 137 mM NaCl, 5.3 mM KCl, 0.4 mM KH2PO4, 4.2 mM NaHCO3 and 0.4 mM Na2HPO4, pH=7.2, supplemented with 10 mM D-glucose) at 4ºC. Cells were resuspended in the same buffer at final density of 4x107 cells/mL or 2x107 cells/mL, and kept on ice until use. 62 Determination of antimicrobial activities Principles The antimicrobial activity of the different peptides was determined by colorimetric measurement of the reduction of MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide) to blue insoluble formazan by mitochondrial reductases. When cell viability is impaired, this leads to a decrease in formazan formation; by the same rule, when tested on parasite proliferation, the amount of formazan is proportional to the number of viable cells. This method can be applied to assess the impact of an antimicrobial peptide on cell viability (short-term effect) and on the cell proliferation (long-term effect), using the same sample, and determine parameters like the IC50 (the concentration of AMP that inhibits 50% of MTT reduction at the end of 4h) and the LD50 (lethal concentration that causes 50% inhibition of parasite proliferation at the end of 3 or 5 days, for promastigotes and amastigotes, respectively) (Luque-Ortega, J.R. & Rivas, L. 2010). Method Parasite suspension (4x107 cells/mL) was distributed (60 μL/well) in a sterile 96-well microplate; 60 μL of the corresponding peptide solution (all the solutions and suspensions in HBSS + 10 mM glucose) was added to the respective well, and incubated for 4h at 26 ºC or 32 ºC, for promastigotes and amastigotes, respectively. Long-term effect: - For promastigotes, after 4h of incubation 20 μL of each well were transferred into a new 96-well microplate with 130 μL/well of RPMI 1640 without phenol red + 10% HIFCS, and incubated at 26 ºC, for 3 days, to allow parasite proliferation. At the end of the incubation, 50 μL of 2 mg/mL MTT solution was added to each well and further incubated at 26 ºC (protected from light) until colour development. Finally, to stop the reaction and to dissolve the blue formazan crystals, 50 μL of SDS 10% was added to each well. Absorbance was read at 600 nm in a microplate reader (Bio-Rad model 680) fitted with a 595 nm filter. - For amastigotes, in the end of the incubation, 20 μL of each well were transferred to a new 96-well microplate with 130 μL/well of M199 + 10% HIFCS, and incubated at 32 ºC, for 5 days, to allow parasite proliferation. Afterwards, the content of each well was transfer into 1.5 mL microcentrifuge tubes, and wells were additionally washed with 100 μL of HBSS. The tubes were centrifuged and washed again with 1 mL of HBSS. The 63 pellet was resuspended in 100 μL of 0.5 mg/mL MTT solution, and the whole content of the tubes was transferred into a new 96-well microplate and MTT reduction carried out as described previously. For the short-term assay, after incubation with the respective peptide, 100 μL of MTT (1 mg/mL) solution were added to each well of the plate incubated for 4h, and proceeded as described before. The parameters IC50 and LD50 were calculated using the SigmaPlot® software, and the results represented as the media of at least two independent experiments (conditions tested in triplicates) with the corresponding standard deviation (SD) indicated. Real-time monitoring of in vivo changes in intracellular levels of ATP in L. donovani promastigotes Principles A strain of L. donovani promastigote, 3-Luc that express a cytoplasmic form of luciferase was used to monitor the levels of intracellular ATP. This assay is based on the reaction, catalyzed by luciferase that converts D-luciferin into D-oxoluciferin in the presence of O2, with consumption of one molecule of ATP and emission of one foton. To overcome the low permeability of the cellular membrane to D-luciferin at physiological pH, a caged ester analogue of luciferin, DMNPE-D-luciferin (D-luciferin, 1-(4,5-dimethoxy-2-nitrophenyl) ethyl ester), free-permeable to the membrane, was used. Once inside the cytoplasm Dluciferin is released by intracellular esterases. In this case, luminescence is directly related to the cytoplasmic levels of free-ATP, the limiting substrate for the reaction (Luque-Ortega, J.R. et al. 2003). When the cells are intact, and in the presence of luciferase substrate, cells emit light. However in the presence of a peptide that induces membrane damages, the loss of ionic gradient or, for larger lesions, leakage of ATP, with subsequent depletion in the cytoplasm, will lead to the decrease of the light emitted (Luque-Ortega, J.R. et al. 2003). Method Parasites, L. donovani 3-Luc promastigotes, were resuspended at 2x107 cells/mL in HBSS + 10 mM glucose, incubated at 26 ºC with DMNPE-D-luciferin (25 μM, final concentration), and 90 μL of the suspension was distributed into a black 96-well microplate. When luminescence reached a plateau, 10 μL of the respective peptide was added (t = 0) and 64 the decay on the luminescence was monitored in a BMG Polarstar Galaxy microwell reader (Öffenburg, Germany) equipped with luminescence setting. Luminescence data were referred as percentage of luminescence corrected with the corresponding values of untreated promastigotes. Plasma membrane permeabilization of L. donovani promastigotes Principles The presence of a peptide may induce lesions into the cytoplasmic membrane large enough to allow entrance of the cationic vital dye, Sytox Green (MW = 600 Da), to which intact cells are impermeable. Once the dye gains access to the intracellular space, it binds to nucleic acids with high-affinity, close to irreversibility in practical terms. This enhances its fluorescence more than 500-fold (Luque-Ortega, J.R. & Rivas, L. 2010). So the increase in fluorescence stands for the severity of the damage to the parasites membranes. Method Two different assays were performed to assess the entrance of the vital dye: - First, the parasites L. donovani promastigotes (90 μL at 2x107 cells/mL) were incubated with Sytox Green (1 μM, final concentration) in HBSS + 10 mM glucose and fluorescence changes were monitored at λex = 504 nm and λem = 524 nm in a BMG Polarstar Galaxy microwell reader (Öffenburg, Germany), immediately after adding 10 μL of the corresponding peptide solution (t = 0). The results were expressed as percentage relative to full permeabilized cells achieved by the addition of 0.1% Triton X-100. - Second, to assess the incorporation of the dye into individual parasites, L. donovani promastigotes and antimicrobial peptides were incubated at the same conditions as described above, except that the Sytox Green was added only after 4h of incubation at 26 ºC, and the fluorescence was measured by cytofluorometry in a Beckman-Coulter FC-500 Analyzer. Maximal permeation was achieved by 15 μM of digitonin in order to preserve parasite shape. 65 AMPs or CPPs: Confocal Microscopy To evaluate whether the peptides enter the cells without permeabilizing the membrane and extrapolate if these peptides had an internal target to account for their lethal action, in other words, whether they behaved or not as Cell-Penetrating Peptides (CPPs) L. donovani promastigotes and L. pifanoi amastigotes (4x107 cells/mL) were incubated with the peptides labeled with FITC (fluorescein isothiocyanate) in HBSS + 10 mM glucose for 4h at 26 ºC and 32 ºC for promastigotes and amastigotes, respectively. After 4h of incubation propidium iodide, PI, (50 μg/mL, final concentration) was added to the cells. A 15 μL drop of each sample was placed in microscope slides previously coated with polylysine (to favour parasite attachment and immobilization) and parasites were observed and photographed in a Confocal Laser Scanning Microscope Leica TCS – SP2 ABOS. Ultrastructural alterations: Transmission Electron Microscopy L. donovani promastigotes or L. pifanoi amastigotes suspension (4x107 cells/mL) in HBSS + 10 mM glucose were incubated with the peptides for 4h at 26 ºC or 32 ºC, for promastigotes and amastigotes, respectively. After the incubation, cells were washed with PBS and resuspended in a 3% glutaraldehyde solution in PBS for 1-2h, for cell fixation, and then incubated with 2.5% osmium tetroxide (OsO4) for about the same time (1-2h), with PBS washes between steps. Cells were dehydrated by successive incubations with increasing concentrations of ethanol: 30% and 50% for 30 min; 70% overnight; 90% and 100% for 30 min. At the end, parasites were incubated with propylene oxide and afterwards, propylene oxide was successively replaced by a mix of resins 3A / 7B (resin A: 6.88 mL of Epon-812 epoxy resin + 11.11 mL of dodecenyl succinic anhydride (DDSA) – resin B: 22 mL of Epon-812 epoxy resin + 19.8 mL of methyl -5-norbornene-2,3dicarboxylic anhydride (NMA)). The protocol consisted in successive incubations with different ratios of a mixture between propylene oxide and the resins: 3:1 and 1:1 for 30 min, 1:3 overnight and 0:1 for 30 min. The inclusion of cells into the resins was completed with a polymerization step (70ºC for 48h), catalyzed by the addition of 1.8% 2,4,6tri(dimethylamino methyl) phenol (DMP30) as accelerator. The samples were distributed into electron microscopy capsules (Size 3, Agar Scientific, Stanted, England) that were trimmed and sliced with an ultramicrotome LKB. Sections were placed into formvardcarbon-coated copper grids and stained with 25 mM lead citrate. The samples were observed and photographed in a transmission electron microscope JEOL-1230. 66 Colocalization of antimicrobial peptides and L. pifanoi amastigotes in peritoneal macrophages: Confocal microscopy Three days (day 0) prior to macrophage harvesting, BALB/c mice were intraperitoneally (i.p) injected with 1 mL of 10% thioglycolate. On the 3rd day, mice were sacrificed by cervical dislocation, and macrophages were obtained by washing the peritoneum twice with warm PBS. 100 000 cells/well were seeded in a Lab-Tek 16-well chamber slide system, with RPMI 1640 + 10% HIFCS. Macrophages were incubated for 2-3h at 37 ºC to allow cell attachment to the glass bottom of the chamber slide. When the macrophages were well attached to the coverslides, the medium was replaced with a suspension of L. pifanoi amastigotes in fresh medium, at a final ratio amastigote:macrophage of 3:1 and incubated for at least 4h at 32 ºC. Next, non phagocytized parasites were removed by washing each well twice with warm PBS, and substituted for new media (RPMI 1640 + 10% HIFCS) with or without peptides at different concentrations. To ascertain whether FITC-labeled peptides, have any privileged subcellular accumulation inside the infected macrophages, including the parasitophorus vacuole where the Leishmania amastigotes grow, macrophages, previously seeded in a Lab-Tek 16-well chamber slide system, were incubated with the labeled peptides overnight. After that, wells were washed twice with warm PBS. The plastic wells, and any trace of silicone gasket, were removed from the slide carefully according to the manufacturer instructions. Before covering the slide with a coverslide, a drop of PBS was placed in the middle of each preparation to avoid cell drying. Finally macrophages were observed and photographed in a Confocal Laser Scanning Microscope Leica TCS – SP5 ABOS. Determination of macrophage viability To measure macrophage viability in the presence of different peptide concentrations, mouse peritoneal macrophages were seeded (100 000 cells/well) into a 96-well plate. After adherence to the plate, macrophages were incubated, for 4h in 100 μL of HBSS + 10 mM glucose with or without peptides at different concentrations. After peptide incubation, 100 μL of 1 mg/mL MTT solution was added to each well, and MTT reduction was measured as described before. 67 2. RESULTS 2.1. Determination of the antimicrobial activity of lactoferrin-derived peptides against L. donovani promastigotes and L. pifanoi amastigotes Lactoferrin-derived peptides, LFcin17-30, LFampin265-284 and LFChimera were all leishmanicidal (determined by MTT reduction) against L. donovani promastigotes below 50 μM (table 5). LFChimera was the most active, with IC50 and LD50 values around 4 μM (table 5). All the peptides were less active against amastigotes than promastigotes. LFcin and LFampin had IC50 and LD50 values above 50 μM against amastigotes while the LFChimera was once again more active, with IC50 and LD50 values below 10 μM (table 5). Leishmania parasites were not able to recover from the damage induced by the peptides after 4h of incubation (represented by IC50), as the LD50 was always similar to the IC50, for both promastigotes and amastigotes (table 5). The inhibitory activity of LFChimera was always higher than the sum of its separated components (figure 21), stressing the importance of having the peptides bond to each other. Table 5 – Leishmanicidal activity of LFcin17-30, LFampin265-284, LFChimera and LFChimera surrogates against L. donovani promastigotes and L. pifanoi amastigotes. The values correspond to the concentration causing 50% inhibition of MTT reduction. Each value represents the average and standard deviation of at least two independent experiments. AMP L. donovani promastigotes L. pifanoi amastigotes IC50 / μM LD50 / μM IC50 / μM LD50 / μM LFcin17-30 21.4 ± 1.3 25.7 ± 1.4 > 50 > 50 LFampin265-284 29.1 ± 1.7 33.2 ± 1.5 > 50 > 50 LFChimera 3.7 ± 0.4 3.6 ± 0.1 6.0 ± 1.1 4.7 ± 1.1 LFChimera-R 3.5 ± 0.1 2.6 ± 0.4 4.2 ± 0.3 3.6 ± 1.1 Di-LFcin 2.4 ± 0.1 2.1 ± 0.2 3.4 ± 0.6 3.2 ± 0.6 Di-LFampin 4.4 ± 0.1 4.0 ± 0.4 4.8 ± 0.8 5.9 ± 1.4 LFcin – LFampin 3.9 ± 0.1 3.4 ± 0.2 4.8 ± 0.6 4.9 ± 1.2 LFampin – LFcin 1.6 ± 0.1 1.6 ± 0.1 1.5 ± 0.3 0.73 ± 0.02 68 0 20 40 60 80 100 120 0246810 % leishmania cell viability [AMPs] / μM LFcin17-30 LFampin265-284 LFChimera LFcin17-30 + LFampin265-284 Figure 21 – Effect of lactoferrin-derived peptides on the viability of L. donovani promastigotes. Parasites were incubated with different concentrations of LFcin17-30, LFampin265-284, LFChimera, and with its two constituent peptides together, for 4h in HBSS + 10 mM glucose. The viability of the parasites was measured by MTT reduction. The results of one representative experiment out of three are shown. In order to determine structure-activity relationships for LFChimera, a set of LFChimera surrogates was tested along with the parental hybrid peptide. In preliminary experiments, the IC50 and LD50 of these peptides against L. donovani promastigotes and L. pifanoi amastigotes were determined by MTT reduction. All the surrogates tested had an activity similar to that of LFChimera, except for two peptides that were even more active, Di-LFcin and LFampin-LFcin (table 5). This last one was active below 2 μM against both forms of the parasite (table 5). An interesting fact was that all peptides had similar activities on promastigotes and on amastigotes, which was not observed for LFChimera. So, the attempt to reach structure-activity relationships is not yet fully accomplished, as the experiments so far were limited to the small number of surrogates available, and further experiments with different surrogates to obtain deeper information about structure-activity relationships are already planned. Nevertheless, some preliminary conclusions can be drawn: i) it seems that the presence of two lactoferricins 17-30 (Di-LFcin) improves the activity of the peptide due to a higher charge (+14 compared to +12 of LFChimera and to +10 of Di-LFampin); ii) the fact that a linear hybrid, LFampin-LFcin, has higher activity than LFChimera suggest that the link between the LFcin and LFampin through a additional lysine is not crucial for the antimicrobial activity of the parental hybrid peptide. Moreover in these linear peptides it seems important the order of the constituent peptides, as LFampin-LFcin was more active than LFcin-LFampin. Due to the finding of at least two very promising peptides, that could be used in the future to fight leishmaniasis, further experiences will be performed with these LFChimera surrogates to uncover their mechanism of action and cytotoxicity. 69 2.2 Evaluation of Leishmania plasma membrane permeabilization by lactoferrinderived peptides In order to test whether the lethal mechanism of lactoferrin-derived peptides against Leishmania correlates with a plasma membrane permeabilization process, a set of permeabilization assays were performed. These experiments can elucidate on how these peptides act, the time-frame of action, what is the main target, among others. First the intracellular levels of free-ATP in living parasites, L. donovani promastigotes, were monitored. Immediately after peptide addition (t=0) there was a fast and dosedependent bioenergetic collapse of the parasites (figure 22). This means that the presence of the peptide disturbs the membrane, impairing the production of ATP by the parasites, due to loss of the ionic gradient or leakage of ATP, leading to their death. Once again the effect of LFChimera was stronger, as for much lower concentrations (between 10 and 1.25 μM) a similar or even higher effect was observed, when compared with the other two peptides. Figure 22 – Real-time depletion of intracellular ATP levels upon peptide addition in living L. donovani 3-Luc promastigotes. Measurement of intracellular levels of ATP, by luminescence, in L. donovani 3-Luc promastigotes after peptide (LFcin17-30, LFampin265-284 and LFChimera) addition at t = 0 (luminescence = 100%). The results of one representative experiment out of two are shown as percentage relative to control parasites (without the addition of AMPs). With the above results, it was recognized that these peptides induce membrane damages. However how severe were those damages? Are they big enough to let a cationic molecule, Sytox Green, with 600 Da to enter inside the cytoplasm? To answer these questions the entrance of Sytox Green was monitored, after adding the peptide, by the fluorescence enhancement, due to the binding of the dye to nucleic acids. The results obtained were similar and in agreement with the luminescence results. It was possible to see a fast and dose-dependent uptake of the dye that reaches saturation minutes after peptide addition 76 77 III.2 – Activity of cecropin A-mellitinderived and lactoferrin-derived peptides against Mycobacterium avium 78 79 1. MATERIAL AND METHODS Peptides In the evaluation of the activity against Mycobacterium avium, the following peptides were used: cecropin A-melittin derivatives CA(1-7)M(2-9) or CAM, K6 and K7 (described in table 3, page 55) and lactoferrin-derived peptides hLF1-11, LFcin17-30, LFampin265-284 and LFChimera (described in table 4, page 56). Bacteria M. avium strain 2447, smooth transparent variant (SmT) was isolated from an AIDS patient and provided to us by Dr. F. Portaels (Institute of Tropical Medicine, Antwerp, Belgium). Mycobacteria were grown to mid-log phase in Middlebrook 7H9 medium (Difco, Sparks, MD) containing 0.05% of Tween 80 (Sigma, St. Louis, MO) and 10% of ADC supplement (Albumin-Dextrose-Catalase) at 37 ºC. Bacteria were harvested by centrifugation, washed twice with saline containing 0.04% Tween 80, resuspended in the same solution and briefly sonicated in order to disrupt bacterial clumps. The suspension was stored in aliquots at -80 ºC until use. Just before use, an aliquot was quickly thawed and diluted to the appropriate concentration. Effect of antimicrobial peptides on the viability of M. avium in axenic cultures Principles To determine the direct effect of the antimicrobial peptides on M. avium 2447 SmT in axenic cultures (bacteria growing in liquid medium), three different methods were used. First the bacterial growth was monitored by optical density at 600 nm. When they were in the exponential phase of growth, the viability of M. avium was quantified by resazurin reduction and by colony forming units (CFUs). Resazurin is a blue oxidation-reduction indicator (Palomino, J.C. et al. 2002) which is nonfluorescent until it is reduced to the pink colored and highly fluorescent resorufin. Resazurin reduction can be used to measure cellular viability, as only viable cells can reduce this dye and the level of resazurin reduction is proportional to the number and viability of the cells present. This is a simple and non-destructive assay, that can be 80 applied both to adherent and non-adherent cells, and it complements and gives comparable data with the colony forming units (CFUs) assay (Anoopkumar-Dukie, S. et al. 2005). Method M. avium 2447 SmT was grown in Middlebrook 7H9 medium to exponential phase. 107 CFU of bacteria were seeded per well into 96-wells flat bottom plates and incubated with the peptides in a final volume of 200 μL. Each peptide concentration was tested in triplicates. The plates were incubated at 37 ºC in a humid atmosphere. At the 6th day of culture, 10 μL of 2.5 mM resazurin was added to each well and after 24h of incubation the fluorescence was measured in a Gemini XM plate fluorometer (Molecular Devices) (λex = 560 nm; λem = 590 nm). The results were expressed as percentage of the fluorescence obtained in control wells without peptides. % mycobacterial cell viability = relative fluorescence intensity in experimental wells relative fluorescence intensity in control wells x 100 In some of the assays, after 7 days of incubation, the viability of the bacteria was also measured by a colony forming units (CFUs) assay (using the same wells as the resazurin assay). The bacterial suspension in each well was serially diluted in water containing 0.04% Tween 80. The dilutions were plated in Middlebrook 7H10 agar medium (Difco, Sparks, MD), supplemented with OADC (Oleic acid-Albumin-Dextrose-Catalase) and the number of colonies was counted after 7 days at 37 ºC. The difference, in terms of log10 CFU/mL between the first and the last day of incubation was designated “log increase”. log increase CFU/mL = log10 CFU/mL (day 7) – log10 CFU/mL (day 0) Collection, infection and treatment of Bone Marrow derived Macrophages (BMMØ) Macrophages were derived from the bone marrow of BALB/c INF-γ knock-out mice. Each femur and tibia was flushed with 5 mL of Hank’s Balanced Salt Solution (HBSS, Gibco, Paisley, U.K.). The cell suspension was centrifuged and resuspended in Dulbecco’s Modified Eagle’s Medium (DMEM, Gibco, Paisley, U.K.) supplemented with 10 mM glutamine, 10 mM HEPES, 1 mM sodium pyruvate, 10% Fetal Bovine Serum (FBS, Gibco, Paisley, U.K.) and 10% of L929 cell conditioned medium (LCCM) as a source of Macrophage Colony Stimulating Factor (M-CSF). 81 Cells were cultured overnight at 37 ºC in a 7% CO2 atmosphere to remove fibroblasts. Non-adherent cells were collected with cold HBSS medium, washed and seeded at the concentration of 5x105 cells/well in 24-wells culture plates (Nunc, Gibco). In some cases LPS-free round glass coverslips were placed at the bottom of culture wells, so that macrophages could adhere to this coverslip and later be observed by confocal microscopy. The plates were incubated at 37 ºC in a 7% CO2 atmosphere. Four days after seeding, 10% of LCCM was added to the culture medium and on the 7th day, the medium was renewed. At day 10 of culture, when the cells are fully differentiated into macrophages, each well was incubated with 106 CFU of M avium 2447 SmT in DMEM, for 4h at 37 ºC in a 7% CO2 atmosphere. After incubation, cells were washed several times with warm HBSS to remove non-internalized bacteria, and re-incubated with new medium, DMEM with 10% FBS and 10% LCCM, with or without different concentrations of peptides alone or in combination with different antibiotics. Each condition was tested in triplicates. After 4 or 7 days of infection, the intracellular growth of M. avium 2447 SmT was evaluated by colony forming units (CFUs). Macrophages were lysed, using 0.1% saponin. The bacterial suspensions were serially diluted in water containing 0.04% Tween 80 and plated in Middlebrook 7H10 agar medium supplemented with OADC. The number of colonies was counted after 7 days at 37 ºC. The difference, in terms of log10 CFU/well between the first and the last day of culture was designated “log increase”. log increase CFU/well = log10 CFU/well (day 4 or 7) – log10 CFU/well (day 0) Determination of macrophage viability Principles The viability of infected macrophages was determined by resazurin reduction (see above) and by the use of VibrantTM Cytotoxicity Assay Kit V-23111 (Molecular Probes) (both assays can be performed using the same wells). The cytotoxicity assay kit monitors the release of the cytosolic enzyme glucose 6phosphate dehydrogenase (G6PD) from damaged cells into the surrounding medium. This enzyme is detected in cell culture supernatants by a two-step enzymatic process that leads to the reduction of resazurin (figure 30), and the resulting fluorescence signal is proportional to the amount of G6PD released which in turn correlates with the number of dead cells in the sample. 82 Figure 30 – Principle of the coupled enzymatic assay for detection of glucose 6-phosphate dehydrogenase activity. Oxidation of glucose 6-phosphate by glucose 6-phosphate dehydrogenase results in the generation of NADPH, which in turn leads to the reduction of resazurin by diaphorase to yield fluorescent resorufin (Molecular Probes). Method Resazurin reduction as a measure of cell viability: After 4 or 7 days of infection and peptide treatment, 100 μL of resazurin 1.25 mM was added to each well. The plates were incubated for 4h at 37 ºC in a 7% CO2 atmosphere. In the end, the fluorescence was measured in a Gemini XM plate fluorometer and expressed as percentage of the fluorescence obtained in control wells (non-treated infected macrophages). % macrophage viability = relative fluorescence intensity in treated infected macrophages relative fluorescence intensity in non-treated infected macrophages x100 VibrantTM Cytotoxicity Assay Kit to measure cell death: After 2 days of infection and peptide treatment 50 μL of the supernatant of each well was transferred into a new 96wells flat bottom plate. An enzymatic mixture was prepared according to manufactor’s instructions and added to the wells. Fluorescence readings (λex = 530 nm; λem = 590 nm) were taken at 5 minutes intervals at 37 ºC. The relative fluorescence intensity obtained for experimental cells (treated infected macrophages) and for fully lysed cells, was corrected by subtracting the values obtained from the untreated infected macrophages. The relative cytotoxicity was determined by dividing the corrected fluorescence intensity of experimental cells by the corrected fluorescence intensity of fully lysed cells. % cytotoxicity = corrected fluorescence intensity of experimental cells corrected fluorescence intensity of fully lysed cells x 100 83 Confocal Microscopy At day 3rd of infection, macrophages that were cultured in glass coverslips, were washed with warm PBS and fixed with 4% paraformaldehyde (Merck) in PBS supplemented with 120 mM sucrose for 30 min at room temperature. For fluorescence quenching, cells were incubated with PBS containing 50 mM of NH4Cl, for 10 min. Cells were washed again with PBS and incubated with Auramine O stain kit (Polysciences, Inc.). Briefly, cells were incubated with a few drops of Auramine O for 15 minutes (without letting the surface dry and protected from light), then with fluorescent decolorizer, for 30-60 seconds, and finally with potassium permanganate for 2 minutes. Between incubations cells were washed with distilled water. Slides were mounted using Mowiol (Polyscience) and sealed with nail polish. Macrophages were observed and photographed in a Laser Scanning Confocal Microscope Leica SP2 AOBS SE (Leica Microsystems, Germany). 84 85 0% 20% 40% 60% 80% 100% 120% 0 6.25 12.5 25 50 100 200 % mycobacterial cell viability [AMPs] / μM CAM K6 K7 0.00 0.50 1.00 1.50 2.00 2.50 1 2 3 log increase CFU/mL ** ****** ** ** ** ** * *** 2. RESULTS 2.1. Cecropin A-melittin-derived peptides 2.1.1. Effect on M. avium viability in axenic cultures Cecropin A-melittin hybrid, CA(1-7)M(2-9) or CAM, and the respective trimethylated derivatives K6 and K7, had a direct effect against M. avium, significantly inhibiting its growth in liquid medium above 50 μM. However, even at four-fold higher concentration (200 μM) the growth inhibition did not reach 50% (figure 31). This direct effect on M. avium viability was determined, in three independent experiments, by resazurin reduction, and confirmed by CFUs. The results obtained by the two techniques are highly concordant. Figure 31 – Effect of cecropin A-melittin-derived peptides on M. avium viability. M. avium was cultured in liquid medium, in the presence of 0, 6.25, 12.5, 25, 50, 100 and 200 μM of CAM, K6 and K7 for 7 days. The viability was measured by A) resazurin reduction, where the results were expressed as percentage of cell viability relative to control wells, and B) colony forming units (CFUs) assay. Statistical analysis was performed using the two-tailed Student’s t-test: * p<0,05; ** p<0,005; *** p<0,0005. The results of one representative experiment out of three are shown. 2.1.2. Effect on M. avium growing inside macrophages and macrophage toxicity Recently, the hemolytic activity of CAM, K6 and K7 was determined (Fernandez-Reyes, M. et al. 2010), and the analogues K6 and K7 showed a significantly lower hemolytic activity compared to the parental hybrid peptide. However, when we tested the same peptides on macrophage cultures infected with M. avium, they revealed a high toxicity, assessed by resazurin reduction (results not shown) and by the use of VibrantTM Citotoxicity Kit Assay A B [CAM] / μM [K6] / μM [K7] / μM 92 0% 30% 60% 90% 120% 150% 0 2.5 5 10 15 % macrophage viability [LFChimera] / μM T4 T7 seem to be healthy as the morphology of the cells, as far as we can see and assess, was not affected (figure 40). Figure 39 – Toxicity of LFChimera on infected BMMØ. After 4 and 7 days of incubation of infected-macrophages with 0, 2.5, 5, 10 and 15 μM of LFChimera, cellular viability was determined by resazurin reduction. The results of one representative experiment out of two are shown and presented as percentage of viable cells relative to non-treated infected wells in each time-point. Figure 40 – Effect of LFChimera on the intramacrophagic growth of M. avium. Macrophages infected with M. avium for 3 days were fixed and mycobacteria were stained with Auramine O. Cells were photographed in a Laser Scanning Confocal Microscope Leica SP2 with final magnification of 630x, and representative pictures are shown. Upper panels show phase contrast and lower panels the corresponding fluorescence image. A) Non-treated infected macrophages. B) Infected macrophages treated with 10 μM LFChimera. C) Higher magnification of one macrophage from picture B. M. avium – Auramine O M. avium – Auramine O M. avium – Auramine O A B C 93 III.3 – Characterization of peptide-lipid interaction by X-ray diffraction 94 95 1. MATERIAL AND METHODS Lipids 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE or 16:0-18:1 PE) lipid was obtained from Avanti Polar Lipids (Alabaster, AL), and used without further purification. Figure 41 – Structural representation of POPE (Avanti Polar Lipids). Peptides To assay the peptide-lipid interaction on model membranes by X-ray diffraction, cecropin A-melittin derivatives (CAM, K6 and K7) and lactoferrin-derived peptides (LFcin17-30, LFampin265-284 and LFChimera) were used (see tables 3 and 4 on pages 55 and 56). Preparation of liposomes To obtain multilamellar vesicles (MLVs) of POPE, 2 methods were used: 1. Appropriate amounts of the lipid were dissolved in a small volume of chloroform, and transferred to a round flask, where a film was obtained by drying the chloroform under a slow nitrogen flow. The film was thereafter kept under vacuum for 3h to remove all traces of organic solvents. The lipid film was hydrated with PBS (150 mM NaCl, pH 7.2) and warmed for about 1h in a thermostated bath at 35 ºC, above the main transition temperature (TM = ~25 ºC) and below the lamellar to hexagonal transition (TH = ~71 ºC) to guarantee that all lipids are in the fluid lamellar state. The resulting vesicles (MLVs) were frozen in liquid nitrogen and thawed in the thermostated bath, and this process was repeated 5 times. 2. Appropriate amounts of the lipid were hydrated with PBS (150 mM NaCl, pH 7.2), and incubated for 30 min in a water bath at ~35 ºC, above the main transition temperature, 96 for the same reasons as stated above. After this, a large number of short cycles of alternating vortex at room temperature and heating at 35 ºC were performed. An aliquot of each liposome suspension was stored at -20 ºC for determination of the exact lipid concentration. The remaining lipid was kept at 4 ºC until use. Peptide-lipid mixture’s preparation Peptide solution in the same buffer used for preparing the liposomes was added to the liposome suspension at different peptide-to-lipid (P:L) ratios, and the mixtures were incubated for 30 min at 35 °C. The samples were then centrifuged at 13 000 rpm in a microcentrifuge at least for 15 min, and transferred thereafter into glass capillaries (Spezialglas Markröhrchen 1.5 mm capillaries, Glass Technik & Konstruktion – Müller & Müller OHG, Germany). The capillaries were sealed by flame, and stored at 4 ºC, at least one day before use. X-Ray diffraction Principles X-ray diffraction is a powerful tool to characterize the structure of lipid phases and polymorphic changes in the lipid bilayer. Lipid dispersions in water form structures with long-range periodic order that can be analyze by X-ray techniques, in particular by diffraction methods, as these periodic structures will diffract the X-rays in a particular pattern depending on the existent phase and the respective space group. This is described by Bragg’s law mλ = 2dsinθ where m is the diffraction order, λ is the wavelength of the diffracted radiation, d is the lattice spacing and 2θ is the scattering angle. Measurements of the intensity of the scattered radiation as a function of the scattering angle allows the determination of the lattice parameter (Hanulóva, M. 2008). The radiation can be diffracted at small angles, up to 3º – Small Angle X-ray Diffraction (SAXD) or in larger angles, up to 20º – Wide Angle X-ray Diffraction (WAXD) (Hanulóva, M. 2008). Therefore, by SAXD we can get information about distances in large structural repetitions that are in the order of 10 to several 100 Å, like the lattice parameter (e.g. 97 bilayer spacing), whereas WAXD provides information about repeating structural patterns at small distances (2 to 9 Å), like the hydrocarbon chain lattice, if they are not in a liquid disordered state. Ideally, these two distinct regions of the diffraction pattern should be measured simultaneously to characterize completely the structure present. Lamellar, hexagonal and cubic phases can be detected and characterized by X-ray diffraction. The first two phases are normally present in microscopic crystalline domains, randomly oriented throughout the sample, whereas cubic phases can grow to large dimensions, in some cases of the size of the sample container. With the exception of the last one (cubic phase), a few diffractions are enough to unequivocally describe the type of symmetry (e.g. lamellar or hexagonal) and to obtain the repeat distances (d or a). When we measure the intensity of the scattered radiation as a function of the scattering angle in the SAXD region (the SAXD pattern), we obtain peaks, which for a gel (Lβ) and a fluid lamellar phase (Lα) are equidistant (figure 1) and appear at sh = h/d (h – Miller index), reflecting the bilayer spacing (or lattice parameter of lamellar phase – d in figure 42). In WAXD a sharp peak appears for the gel lamellar phase (Lβ) and only a broad diffuse peak for the liquid lamellar phase, as in the gel lamellar phase the lipid chains are more extended and rigid, and thus have order reflections and thus this can be used to safely assess the presence of a gel lamellar phase (Hanulóva, M. 2008). Figure 42 – SAXD pattern of the fluid lamellar phase (Lα) of POPE membranes. The corresponding structure (adapted from: Tresset, G. 2009), is shown in the inset with the respective lattice parameter (d). 98 The diffraction pattern of a hexagonal phase (HII) in the SAXD region contains a series of peaks that appear at positions shk =(2 h2+k2+hk 3ah) (h, k – Miller indices), where ah is the distance between the centers of the neighboring lipid cylinders (figure 43). In WAXD, similar to the fluid lamellar phase, only a broad peak appears (Hanulóva, M. 2008). Figure 43 – SAXD pattern of the inverse hexagonal phase (HII) of POPE membranes. The corresponding structure (adapted from: Tresset, G. 2009), is shown in the inset with the respective lattice parameter (ah). The SAXD diffraction pattern of a cubic phase shows a number of peaks at positions given by shkl = ( h2+ k2+ l2 / ac) (h, k, l – Miller indices), where ac is the cubic lattice parameter (figure 44). The more intense peaks of the cubic phase appear at small s values (reciprocal spacing), reflecting the large repeating units of this phase. In the WAXD region a broad diffuse peak is seen (Hanulóva, M. 2008). Figure 44 – SAXD pattern of the cubic phase Pn3m. The corresponding structure is shown in the inset with the respective lattice parameter (ac) (Hanulóva, M. 2008). 99 Typically, depending on the lipid and temperature, liposomes can form lamellar or hexagonal phases. However the presence of an antimicrobial peptide can induce structural changes, either altering only the lattice parameters or leading to the formation of cubic phases. Understanding the type of polymorphic structure that is present in a peptide-lipid mixture can lead to a deeper knowledge of the mechanism of action of the peptide (Haney, E.F. et al. 2010). Method SAXD and WAXD experiments were performed at the synchrotron soft condensed matter beamline A2 in HASYLAB at Deutsches Elektronen Synchrotron (DESY), Hamburg, Germany, using a monochromatic radiation of λ = 0.15 nm wavelength. Figure 45 – Beamline A2 in Hasylab. A) A general view of the instrument is shown. B) In detail, the capillary holder and the two detectors for SAXD and WAXD. Up and down temperature scans were performed for each sample, all at a scanning rate of 1 ºC/min, over the temperature range 20 - 80 ºC. Diffractograms were taken with 10 s of exposure time at each temperature. The heating and cooling of the sample was regulated by a thermocouple connected to the temperature controller JUMO IMAGO 500 (JUMO GmbH & Co. KG, Fulda, Germany). Each SAXD diffraction pattern is presented as normalized diffraction intensity in arbitrary units versus the reciprocal spacing s. The SAXD patterns were analysed in two different ways: i) by use of the A2tool program as a first approach to the (qualitative) analysis of the results; and ii) quantitatively by analyzing each difractogram in detail, where the diffraction peaks were fitted by Lorentzians above a linear background by use of the Origin software, in order to derive the lattice parameters (d or a) that characterize the different phases present. The positions of the peaks plotted vs. h for a lamellar phase, h2+hk+k2 for a hexagonal phase or h2+k2+l2 for a cubic b be ea am m d di ir re ec ct ti io on n c ca ap pi il ll la ar ry y h ho ol ld de er r S SA AX XD D W WA AX XD D 100 phase (h, k, l – Miller indices), should fall on a straight line, passing through the origin (0,0), and the lattice parameter d, ah or ac, for lamellar, hexagonal and cubic phase, respectively, is given by the inverse of the slope. WAXD measurements were used to verify the existence of the gel phase (Lβ) (not shown). 101 2. RESULTS 2.1. Cecropin A-melittin-derived peptides 2.1.1. Phase behavior of POPE membranes The diffraction profile of a lipid changes with temperature. In the case of pure POPE membranes, as the temperature increases there is a transition from gel lamellar phase (Lβ) to fluid lamellar phase (Lα) at 26 C, and from fluid lamellar (Lα) to hexagonal phase (HII) at 71 C (figure 47). When the peptides (CAM, K6 or K7) were added to the lipid system, the transition from gel to fluid lamellar phase was not significantly affected, neither in terms of transition temperature nor in lattice value, but some changes were observed in the transition from fluid lamellar to hexagonal phase. This difference was already apparent in the temperature scan of each peptide/lipid mixture (figure 46) and later was confirmed by the determination of the lattice parameters (figure 47). The studied peptides have somewhat different effects: 1. CAM does not alter the transition temperature to hexagonal phase. 2. For both K6 and K7 we observed the coexistence of hexagonal phase (figure 46 – green arrows) together with fluid lamellar phase (figure 46 – red arrows), from about 10 C below the TH of pure POPE transition. This suggests that these two peptides destabilize the lamellar phase and have better compatibility with the hexagonal phase, as it appears at lower temperatures than in the absence of the peptide. Thus, part of the mixture moves to the hexagonal phase (peptide/POPE) earlier, whereas the remaining, presumably not affected by the peptide, remains in the lamellar phase. After the transition to hexagonal phase, it is possible to observe, for the three peptides, the superimposition of two hexagonal phases with close periodicity (figure 47). Similarly to the early appearance of the hexagonal phase, these two hexagonal phases are most probably due to the existence of distinct domains within the membrane - the peptides bind and destabilize only part of the membrane, this being reflected in a different lattice spacing of one hexagonal phase as compared to the one composed of pure lipid. For instance, at 80 ºC the lattice parameter of the hexagonal phase for pure POPE is 7.09 nm, whereas in the presence of CAM, one of the hexagonal phases has a value of 7.12 nm (similar to the pure lipid) and the other 6.89 nm, the same happening for the other two peptides (K6 – 7.11 and 6.85 nm; K7 – 7.10 and 6.88 nm). 108 y = 0.1918x R² = 1 y = 0.0743x R² = 0.9999 y = 0.0614x R² = 0.9994 0.00 0.10 0.20 0.30 0.40 0.50 0 1 2 3 4 5 s / nm-1 (h2+ k2+ l2)1/2 ▲Lα Pn3m Im3m y = 0.1922x R² = 1 y = 0.0714x R² = 0.9998 y = 0.0553x R² = 0.9997 0.00 0.10 0.20 0.30 0.40 0.50 0 1 2 3 4 5 s / nm-1 (h2+ k2+ l2)1/2 ▲Lα Pn3m Im3m y = 0.1917x R² = 1 y = 0.0751x R² = 0.9998 y = 0.0616x R² = 0.999 0.00 0.10 0.20 0.30 0.40 0.50 0 1 2 3 4 5 s / nm-1 (h2+ k2+ l2)1/2 ▲Lα Pn3m Im3m Figure 52 – Cubic phases (Im3m and Pn3m) in POPE membranes with lactoferrin-derived peptides. SAXD patterns for peptide/lipid mixtures at 50 ºC on cooling. A) LFcin17-30 at PL=1:8, B) LFampin265-284 at PL=1:8 and C) LFChimera at PL=1:18. Insert: A plot of s = f( h2+k2+l2 ) for all observed reflections is shown. A B C 109 PART IV DISCUSSION & CONCLUSIONS 110 111 DISCUSSION & CONCLUSIONS In this work we used microbiological and biophysical techniques to study the effects of cecropin A-melittin derived and lactoferrin-derived peptides, on different pathogens and on model membranes. The antimicrobial activity of cecropin A-melittin derivatives was assessed against Mycobacterium avium in axenic cultures (bacteria growing in liquid medium) and in cultures of macrophages, the cell inside which mycobacteria usually grow. CAM, K6 and K7 directly inhibit M. avium growth above 50 μM but even at 200 μM the growth inhibition did not reach 50%. This lack of significant effect suggests that these peptides are not able to crossover the highly impermeable and complex mycobacterial cell wall. In addition, these peptides are toxic towards macrophages at much lower concentrations, the cytotoxicity of CAM and K6 at 20 μM overpassing 50%. For concentrations at which the toxicity does not reach 10%, no effect on the viability of M. avium growing inside macrophages was seen. All these results discourage the use of these particular peptides as therapeutic agents against mycobacterial infections, as their activity was very low and they cannot be used on macrophages, the host cell of these bacteria. Nevertheless the results are interesting in some points of view. For instance, recent studies showed that the hemolytic activity of the trimethylated peptides K6 and K7 was very low (Fernandez-Reyes, M. et al. 2010), which was seen as a promising characteristic in terms of therapeutic utility. However, our results showing that the peptides were toxic towards macrophages, indicate that although erythrocytes and macrophages are both mammalian cells, there must be crucial differences in their cellular membranes. This suggests that the use of erythrocytes only as target cells for determination of AMP toxicity may be misleading. The differential effects of cecropin A-mellitin peptides on different cell types could be more clarified with biophysical studies using model membranes mimicking these different cells. Until now the most common model membrane of choice for mammalian cells is solely composed by phospatidylcholine (PC) the main lipid of the cellular membrane of erythrocytes. The lipid composition of the cytoplasmic membrane of macrophages must be very different from the membrane of erythrocytes, containing other phospholipids and possibly changing the relative amounts, and this could explain the differences in behavior that we encountered. However the knowledge about the lipid composition of the macrophage cytoplasmic membrane is poor and so no model membranes more specific for these cells are known up to now. The use of lipidomics, for instance, could be useful to try to uncover this composition, and be subsequently used to make model membranes, leading to a deeper 112 knowledge on how these peptides, and others, interact with these types of cells, and in that way refine the applicability of the biophysical techniques as screening methods for the antimicrobial activity of the AMPs. We hope to use this in the future. We have also investigated the effects of several lactoferrin-derived peptides on two intramacrophagic pathogens, Leishmania and Mycobacterium avium. In this study, we found that Lactoferricin (LFcin17-30), lactoferrampin (LFampin265-284) and LFChimera were active against L. donovani promastigotes, and LFChimera is of special interest due to its high activity both against L. donovani promastigotes and L. pifanoi amastigotes. Our results indicate that the leishmanicidal mechanism of these AMPs consists on targeting the plasma membrane, leading to its permeabilization that culminates into bioenergetic collapse of the parasites due to loss of the ionic gradient and leakage of ATP, and also leakage of the cytoplasmic content, ending in parasitic death. The mechanism of action of LFcin17-30 was also very interesting and peculiar, as it appears to induce cellular aggregation. The fact that this peptide was able to reach intracellular amastigotes when they were inside its host cell strongly suggests that it will also be able to exert its leishmanicidal effect inside macrophages. However, the effect of this peptide on the parasitic load inside macrophages was not determined. Future experiments will be performed to address this question. The activity of LFChimera against L. donovani promastigotes was much higher than the sum of its separated components, revealing that having the two peptides, LFcin17-30 and LFampin265-284, bond to each other is important for the antimicrobial activity. To try to understand what are the special features in LFChimera that make this peptide so active, a set of surrogates that varied some of the characteristics of this peptide, such as charge, order of the constituent peptides (LFcin17-30 and LFampin265-284), and the bond between them, were also tested. The activity against Leishmania parasites of all the peptides was equal or higher than the parental peptide. Two of these peptides, Di-LFcin and especially LFampin-LFcin, were found to have higher activity than LFChimera, against both forms of the parasite, promastigotes and amastigotes. This is in contrast with the LFChimera, which was less active in amastigotes as compared to promastigotes. These results suggest that the charge (higher in Di-LFcin in comparison to LFChimera) is important for the activity of AMPs and that the bond between the peptides appears to play an important role too, as in the most potent surrogate (LFampin-LFcin) instead of having the peptides linked by an additional lysine in a special linkage through the side chain (like in LFChimera), they are linked trough a normal peptide bond. So, the attempt to reach structure-activity relationships is not yet fully accomplished, as the experiments so far 113 were limited to the small number of surrogates available, and so, further studies will be done with different surrogates to obtain deeper information about structure-activity relationships. Nevertheless we should stress that at least two very promising peptides were found, with high antimicrobial activity that makes them good candidates to fight leishmaniasis. Further studies should be done in order to understand what are the antimicrobial mechanisms of these peptides. Also of importance, their toxicity towards mammalian cells, such as macrophages and erythrocytes, should be carefully evaluated. Lactoferrin-derived peptides had variable impact on the growth of M. avium. Lactoferrampin had no effect against M. avium either in axenic cultures or growing inside macrophages. Lactoferricins (human, hLFcin1-11, and bovine, LFcin17-30, peptide) significantly inhibited the growth of M. avium axenically at and above 25 μM. Although they did not inhibit the growth of M. avium inside macrophages up to 40 μM, they showed a synergistic effect with ethambutol in this model. These results suggest that the peptide is increasing the access of the antibiotic to the mycobacteria that are growing inside the phagosome of the macrophage. This synergistic effect of lactoferricin with conventional antibiotics was already shown for several other pathogens (Gifford, J.L. et al. 2005; Jenssen, H. & Hancock, R.E. 2009; Leon-Sicairos, N. et al. 2006; Oo, T.Z. et al. 2010; Sanchez-Gomez, S. et al. 2011). It is thought that this peptide disturbs the membrane, in this case the cytoplasmic membrane of the macrophage and/ or the phagosome membrane, probably by forming transient pores increasing the permeability of the membrane not only to the peptide itself but also to compounds that are close to these pores, such as the antibiotic. The combined efficacy could also be due to a more localized effect on the mycobacteria cell wall. The peptide could induce dissipation of the protonmotive force resulting in decreased activity of ATP-dependent multidrug efflux pumps, allowing in that way the entrance of the antibiotic (Gifford, J.L. et al. 2005). Nevertheless, whatever the mechanisms are, the combination of lactoferricin with antibiotics like ethambutol, for the treatment of mycobacterial infections, as for others, has many advantages such as the decrease in the antibiotic dosages and in the duration of the treatment, and more important the decreased ability of the bacteria to acquire resistance to both AMPs and antibiotics. LFChimera, contrary to the effect observed on Leishmania and on other pathogens (Bolscher, J.G.M. et al. 2009; Flores-Villasenor, H. et al. 2010; Leon-Sicairos, N. et al. 2009; Lopez-Soto, F. et al. 2010), did not inhibit the growth of M. avium. In fact the number of M. avium growing inside macrophages increased almost two-fold in the presence of LFChimera at 10 μM. This result was found to be dose-dependent and statistically significant as early as 4 days after infection, with no apparent impact on 114 macrophage viability. Indeed the macrophage viability after 7 days of infection and incubation with LFChimera was maintained at somewhat higher levels than in untreated macrophages. These unexpected results raise the question of the mechanisms underlying the increased growth of M. avium. We can try to propose some explanations. This peptide could be: i) inhibiting a mechanism that normally contributes for bacterial growth restriction (like production of reactive oxygen and/ or nitrogen species); ii) facilitating the access of the bacteria to some critical nutrient or growth factor; iii) interfering with lipid metabolism and iron availability; iv) interfering with the characteristics of the membranes of mycobacteria-containing phagosomes, since AMPs act on membranes, and so many more. Understanding the processes driving this effect could help to clarify the mechanisms by which macrophages actually control the intracellular growth of mycobacteria, leading to the discovery of possible new targets for antimicrobial agents. Also somewhat conflicting are other results revealing that LFChimera was toxic towards peritoneal macrophages infected with L. pifanoi amastigotes. This difference in toxicity between mycobacteria-infected BMMØ and leishmania-infected-peritoneal macrophages could result from: - the peptide-to-cell ratio. Although the concentrations of peptide was the same in both assays the amount of peritoneal cells was 5 times less than the amount of bone marrow derived macrophages (BMMØ) used in the mycobacterial infections, so the number of peptide molecules per cell is much higher in the case of peritoneal macrophages. - the type of medium. The peritoneal macrophages were kept in HBSS buffer supplemented with D-glucose 10 mM, whereas BMMØ were cultured in a complete medium with 10% of fetal bovine serum that can potentially decrease the toxicity of the peptide. - peritoneal macrophages and BMMØ are primary macrophages but the first ones are more differentiated, with less capacity to maintain viability in culture, potentially being more susceptible to the activity of the peptides. So the comparison of results obtained with different techniques and different cells should be done very carefully and every variable should be considered before taking any definitive conclusion. In an attempt to shed some light on the mechanism of action of these families of peptides, the interaction between AMPs and simple model membranes was characterized in this work by a biophysical technique, SAXD. 115 The choice of the lipid composition of the model membranes is important when we want to use biophysical results to help interpret the biological action. As such, we have used mainly model membranes of phosphatidylethanolamine (PE), as it is the most abundant phospholipid of the bacterial membranes. Studies were also performed with lipid mixtures composed of PE and phosphatidylglicerol (PG), as PG is also one of the major phospholipids in the bacteria cell membrane, being one of the main responsible for its characteristic negative charge. Nevertheless, these results are still being analyzed. Studying the phase behavior of the model membranes in the absence and presence of AMPs can provide information about their mechanism of action. By SAXD we can retrieve this structural information by studying the lamellar/non-lamellar phase transitions and the changes induced by the presence of the peptide. In the last years growing evidence is showing that AMPs are able to induce or favor the formation of non-lamellar phases, namelly three dimensional cubic phases, as part of their mechanism of action (Bastos, M. et al. 2011; Haney, E.F. et al. 2010; Zweytick, D. et al. 2008). The formation of such cubic phases, could indicate that the peptides act by promoting an extensive curvature of the bilayer, leading eventually to disruption of the membrane and leakage of the cell content and thereby cell death. Also the existence of cubic phase could indicate that the peptide is inducing cubic domains in the membranes compatible with the formation of toroidal pores, a proposed mechanism for the action of AMPs. All the peptides analyzed induce the formation of bicontinuous cubic phases in POPE liposomes (Pn3m and Im3m) on the cooling process. However each family of peptides affects differently the transition temperatures from lamellar (Lα) to hexagonal phase (HII), in the heating process. In the cecropin A-melittin family all peptides induce changes in the hexagonal phase, either by the existence of two hexagonal phases superimposed or, in the case of the trimethylated peptides (K6 and K7), by the early appearance of the hexagonal phase (10ºC bellow the temperature observed for the pure lipid system), whereas in the case of lactoferrin-derived peptides no changes were observed in this transition (Lα → HII). The main conclusion at this point is thus that the interaction strongly depends on the peptides, and even in the same peptide family quite different patterns of interaction can be observed (e.g. trimethylated peptides, K6 and K7). It should be noted that the behavior also depends on lipid composition – in a recent study (Bastos, M. et al. 2011) we found that lactoferrampin (LFampin265-284) induces the formation of a micellar cubic phase (Pm3n) in model membranes of PC/PG, whereas in the case of PE membranes the same peptide induces the formation of bicontinuous cubic phases. So the interaction between AMPs and lipid membranes is not as unspecific as it was thought some years ago, the 116 details of the interaction go beyond the plain existence of charge at the surface of the bilayer. Although this is work still in progress, we can clearly say at this point that the use of biophysical techniques can help to define the importance of a lipid composition in the overall mechanism of action of antimicrobial peptides and predict their antimicrobial activity. To follow this work, further experiments with lactoferrin-derived peptides are warranted. The potential of LFcin17-30, as well as lactoferricin variants (to be designed de novo) for synergism with other antibiotics, such as rifampicin and clarithromycin, will be assessed against M. avium growing inside macrophages. The mechanism by which LFChimera promotes the growth of M. avium inside macrophages will be assessed by: - infecting macrophages with different intracellular pathogens in order to determine if the effect observed for M. avium is exclusive for this bacteria; - activating macrophages with LPS and IFN-γ, for example, and evaluate the effect of LFChimera on the levels of macrophage antimicrobial molecules (e.g. NO); - infecting phoxand iNOS-deficient macrophages with M. avium and treating with LFChimera to ask whether the growth-promoting effect of the peptide depends in any way on oxygen and nitrogen reactive species; - characterizing LFChimera’s distribution inside non-infected and infected macrophages and the ability of the peptide to reach M. avium containing phagosomes by confocal microscopy using fluorescence-labeled LFChimera; - determining the impact of LFChimera on the degree of maturation of M. aviumcontaining phagosomes by confocal microscopy, using appropriate fluorescent probes and molecular markers; - assessing the integrity of phagosomal and other macrophage membranes when treated with LFChimera by electron microscopy; - determining the effect of LFChimera surrogates on M. avium growing inside macrophages, trying to determine which features of LFChimera are important for the mechanism of promoting the intracellular growth of M. avium. The interaction of these peptides with simple model membranes prepared with different combinations of PC, PG, PE, LPS, phosphatidylserine and cardiolipin, will be assess by Differential Scanning Calorimetry (DSC) and by X-ray diffraction (SAXD and WAXD). More complex preparations of model membranes with extracts of bacteria and extracts of 117 macrophage membranes will also be used, trying to show distinctive patterns characteristic of each peptide/ cell interaction. As a final note we would like to stress that although no immediate or straightforward correlation is apparent between the microbiological and biophysical results, it has been the aim of the groups of Salomé Gomes and Margarida Bastos to perform these studies in parallel, as we have been getting input from the results in one area to lead to changes in procedure or way of “reasoning” in the other. The biophysical insight can help understanding some features of microbiological action, and conversely the later results direct the way that future biophysical experiments are designed. 124 KIMA, P.E. 2007. The amastigote forms of Leishmania are experts at exploiting host cell processes to establish infection and persist. Int J Parasitol. 37(10): 1087-1096. KOBAYASHI, S., CHIKUSHI, A., TOUGU, S., IMURA, Y., NISHIDA, M., YANO, Y. & MATSUZAKI, K. 2004. Membrane Translocation Mechanism of the Antimicrobial Peptide Buforin 2. Biochemistry. 43(49): 15610-15616. KOBAYASHI, S., TAKESHIMA, K., PARK, C.B., KIM, S.C. & MATSUZAKI, K. 2000. Interactions of the novel antimicrobial peptide buforin 2 with lipid bilayers: proline as a translocation promoting factor. Biochemistry. 39(29): 8648-8654. KOPRIVNJAK, T. & PESCHEL, A. 2011. Bacterial resistance mechanisms against host defense peptides. Cell Mol Life Sci. 68(13): 2243-2254. KOUL, A., ARNOULT, E., LOUNIS, N., GUILLEMONT, J. & ANDRIES, K. 2011. The challenge of new drug discovery for tuberculosis. Nature. 469(7331): 483-490. LEON-SICAIROS, N., CANIZALEZ-ROMAN, A., DE LA GARZA, M., REYES-LOPEZ, M., ZAZUETA-BELTRAN, J., NAZMI, K., GOMEZ-GIL, B. & BOLSCHER, J.G. 2009. Bactericidal effect of lactoferrin and lactoferrin chimera against halophilic Vibrio parahaemolyticus. Biochimie. 91(1): 133-140. LEON-SICAIROS, N., REYES-LOPEZ, M., ORDAZ-PICHARDO, C. & DE LA GARZA, M. 2006. Microbicidal action of lactoferrin and lactoferricin and their synergistic effect with metronidazole in Entamoeba histolytica. Biochem Cell Biol. 84(3): 327-336. LODGE, R. & DESCOTEAUX, A. 2005. Modulation of phagolysosome biogenesis by the lipophosphoglycan of Leishmania. Clin Immunol. 114(3): 256-265. LOHNER, K. 2009. New strategies for novel antibiotics: peptides targeting bacterial cell membranes. Gen Physiol Biophys. 28(2): 105-116. LOHNER, K. & BLONDELLE, S.E. 2005. Molecular mechanisms of membrane perturbation by antimicrobial peptides and the use of biophysical studies in the design of novel peptide antibiotics. Comb Chem High Throughput Screen. 8(3): 241-256. LOPEZ-SOTO, F., LEON-SICAIROS, N., NAZMI, K., BOLSCHER, J.G. & DE LA GARZA, M. 2010. Microbicidal effect of the lactoferrin peptides lactoferricin17-30, 125 lactoferrampin265-284, and lactoferrin chimera on the parasite Entamoeba histolytica. Biometals. 23(3): 563-568. LUQUE-ORTEGA, J.R. & RIVAS, L. 2010. Characterization of the leishmanicidal activity of antimicrobial peptides. Methods Mol Biol. 618: 393-420. LUQUE-ORTEGA, J.R., RIVERO-LEZCANO, O.M., CROFT, S.L. & RIVAS, L. 2001. In Vivo Monitoring of Intracellular ATP Levels in Leishmania donovani Promastigotes as a Rapid Method To Screen Drugs Targeting Bioenergetic Metabolism. Antimicrob Agents Chemother. 45(4): 1121-1125. LUQUE-ORTEGA, J.R., SAUGAR, J.M., CHIVA, C., ANDREU, D. & RIVAS, L. 2003. Identification of new leishmanicidal peptide lead structures by automated real-time monitoring of changes in intracellular ATP. Biochem. J. 375(1): 221-230. LUQUE-ORTEGA, J.R., VAN'T HOF, W., VEERMAN, E.C., SAUGAR, J.M. & RIVAS, L. 2008. Human antimicrobial peptide histatin 5 is a cell-penetrating peptide targeting mitochondrial ATP synthesis in Leishmania. FASEB J. 22(6): 1817-1828. LUZZATI, V. 1997. Biological significance of lipid polymorphism: the cubic phases. Curr Opin Struct Biol. 7(5): 661-668. LUZZATI, V., DELACROIX, H., GULIK, A., GULIK-KRZYWICKI, T., MARIANI, P. & VARGAS, R. 1997. Chapter 1 The Cubic Phases of Lipids. Curr Top Membr. 44: 3-14, 14a, 14b, 15-24. MADER, J.S., RICHARDSON, A., SALSMAN, J., TOP, D., DE ANTUENO, R., DUNCAN, R. & HOSKIN, D.W. 2007. Bovine lactoferricin causes apoptosis in Jurkat Tleukemia cells by sequential permeabilization of the cell membrane and targeting of mitochondria. Exp Cell Res. 313(12): 2634-2650. MENDEZ-SAMPERIO, P. 2008. Role of antimicrobial peptides in host defense against mycobacterial infections. Peptides. 29(10): 1836-1841. MERRIFIELD, E.L., MITCHELL, S.A., UBACH, J., BOMAN, H.G., ANDREU, D. & MERRIFIELD, R.B. 1995. D-enantiomers of 15-residue cecropin A-melittin hybrids. Int J Pept Protein Res. 46(3-4): 214-220. 126 Molecular Probes. Vybrant(TM) Cytotoxicity Assay Kit (V-23111). Product information. Available on: http://probes.invitrogen.com/media/pis/mp23111.pdf. Access on: 21July 2011. MOR, A. 2009. Multifunctional host defense peptides: antiparasitic activities. FEBS J. 276(22): 6474-6482. MORGAN, G.W., HALL, B.S., DENNY, P.W., CARRINGTON, M. & FIELD, M.C. 2002. The kinetoplastida endocytic apparatus. Part I: a dynamic system for nutrition and evasion of host defences. Trends Parasitol. 18(11): 491-496. NGUYEN, L.T., HANEY, E.F. & VOGEL, H.J. 2011. The expanding scope of antimicrobial peptide structures and their modes of action. Trends Biotechnol. In Press, Corrected Proof. NICOLAS, P. 2009. Multifunctional host defense peptides: intracellular-targeting antimicrobial peptides. FEBS J. 276(22): 6483-6496. NIGOU, J., GILLERON, M. & PUZO, G. 2003. Lipoarabinomannans: from structure to biosynthesis. Biochimie. 85(1-2): 153-166. OGATA, K., LINZER, B.A., ZUBERI, R.I., GANZ, T., LEHRER, R.I. & CATANZARO, A. 1992. Activity of defensins from human neutrophilic granulocytes against Mycobacterium avium-Mycobacterium intracellulare. Infect. Immun. 60(11): 47204725. OLIVIER, M., GREGORY, D.J. & FORGET, G. 2005. Subversion Mechanisms by Which Leishmania Parasites Can Escape the Host Immune Response: a Signaling Point of View. Clin. Microbiol. Rev. 18(2): 293-305. OO, T.Z., COLE, N., GARTHWAITE, L., WILLCOX, M.D. & ZHU, H. 2010. Evaluation of synergistic activity of bovine lactoferricin with antibiotics in corneal infection. J Antimicrob Chemother. 65(6): 1243-1251. OVERATH, P., STIERHOF, Y.-D. & WIESE, M. 1997. Endocytosis and secretion in trypanosomatid parasites - Tumultuous traffic in a pocket. Trends Cell Biol. 7(1): 27-33. PALOMINO, J.C., MARTIN, A., CAMACHO, M., GUERRA, H., SWINGS, J. & PORTAELS, F. 2002. Resazurin microtiter assay plate: simple and inexpensive method for 127 detection of drug resistance in Mycobacterium tuberculosis. Antimicrob Agents Chemother. 46(8): 2720-2722. PIERCE, A. & LEGRAND, D. 2009. Advances in lactoferrin research. Biochimie. 91(1): 12. PIETERS, J. 2001. Entry and survival of pathogenic mycobacteria in macrophages. Microbes Infect. 3(3): 249-255. PISCOPO, T.V. & MALLIA AZZOPARDI, C. 2007. Leishmaniasis. Postgrad Med J. 83(976): 649-657. Polysciences Inc. Auramine O Stain Kit. Techinical Data Sheet. Available on: http://www.polysciences.com/SiteData/poly/Assets/DataSheets/751.pdf. Access on: 21 July 2011. POZO NAVAS, B., LOHNER, K., DEUTSCH, G., SEVCSIK, E., RISKE, K.A., DIMOVA, R., GARIDEL, P. & PABST, G. 2005. Composition dependence of vesicle morphology and mixing properties in a bacterial model membrane system. Biochim Biophys Acta. 1716(1): 40-48. PRESCOTT, L.M., HARLEY, J.P. & KLEIN, D.A. 2002. Microbiology. Boston, McGraw Hill. QIU, H. & CAFFREY, M. 2000. The phase diagram of the monoolein/water system: metastability and equilibrium aspects. Biomaterials. 21(3): 223-234. RAGHURAMAN, H. & CHATTOPADHYAY, A. 2007. Melittin: a membrane-active peptide with diverse functions. Biosci Rep. 27(4-5): 189-223. RAMALINGAM, K. & BELLO, J. 1992. Effect of permethylation on the haemolytic activity of melittin. Biochem J. 284 ( Pt 3): 663-665. RAPPOLT, M., HICKEL, A., BRINGEZU, F. & LOHNER, K. 2003. Mechanism of the lamellar/inverse hexagonal phase transition examined by high resolution x-ray diffraction. Biophys J. 84(5): 3111-3122. REDDY, K.V., YEDERY, R.D. & ARANHA, C. 2004. Antimicrobial peptides: premises and promises. Int J Antimicrob Agents. 24(6): 536-547. 128 REED, C., VON REYN, C.F., CHAMBLEE, S., ELLERBROCK, T.V., JOHNSON, J.W., MARSH, B.J., JOHNSON, L.S., TRENSCHEL, R.J. & HORSBURGH, C.R. 2006. Environmental Risk Factors for Infection with Mycobacterium avium Complex. Am J Epidemiol. 164(1): 32-40. RIVAS, L., LUQUE-ORTEGA, J.R. & ANDREU, D. 2009. Amphibian antimicrobial peptides and Protozoa: Lessons from parasites. Biochim Biophys Acta. 1788(8): 1570-1581. ROTEM, S. & MOR, A. 2009. Antimicrobial peptide mimics for improved therapeutic properties. Biochim Biophys Acta. 1788(8): 1582-1592. SACKS, D. & NOBEN-TRAUTH, N. 2002. The immunology of susceptibility and resistance to leishmania major in mice. Nat Rev Immunol. 2(11): 845. SANCHEZ-GOMEZ, S., JAPELJ, B., JERALA, R., MORIYON, I., FERNANDEZ ALONSO, M., LEIVA, J., BLONDELLE, S.E., ANDRA, J., BRANDENBURG, K., LOHNER, K. & MARTINEZ DE TEJADA, G. 2011. Structural Features Governing the Activity of Lactoferricin-Derived Peptides That Act in Synergy with Antibiotics against Pseudomonas aeruginosa In Vitro and In Vivo. Antimicrob Agents Chemother. 55(1): 218-228. SHAI, Y., MAKOVITZKY, A. & AVRAHAMI, D. 2006. Host defense peptides and lipopeptides: modes of action and potential candidates for the treatment of bacterial and fungal infections. Curr Protein Pept Sci. 7(6): 479-486. SHARMA, S., VERMA, I. & KHULLER, G.K. 1999. Biochemical interaction of human neutrophil peptide-1 with Mycobacterium tuberculosis H37Ra. Arch Microbiol. 171(5): 338-342. SHARMA, S., VERMA, I. & KHULLER, G.K. 2000. Antibacterial activity of human neutrophil peptide-1 against Mycobacterium tuberculosis H37Rv: in vitro and ex vivo study. Eur Respir J. 16(1): 112-117. SPRONG, H., VAN DER SLUIJS, P. & VAN MEER, G. 2001. How proteins move lipids and lipids move proteins. Nat Rev Mol Cell Biol. 2(7): 504-513. TEIXEIRA, V., FEIO, M.J., RIVAS, L., DE LA TORRE, B.G., ANDREU, D., COUTINHO, A. & BASTOS, M. 2010. Influence of Lysine Nε-Trimethylation and Lipid Composition 129 on the Membrane Activity of the Cecropin A-Melittin Hybrid Peptide CA(1−7)M(2−9). J Phys Chem B. 114(49): 16198-16208. TORTOLI, E. 2006. The new mycobacteria: an update. FEMS Immunol Med Microbiol. 48(2): 159-178. TORTOLI, E. 2009. Clinical manifestations of nontuberculous mycobacteria infections. Clin Microbiol Infect. 15(10): 906-910. TRESSET, G. 2009. The multiple faces of self-assembled lipidic systems. PMC Biophys. 2(1): 3. VAN 'T HOF, W., VEERMAN, E.C., HELMERHORST, E.J. & AMERONGEN, A.V. 2001. Antimicrobial peptides: properties and applicability. Biol Chem. 382(4): 597-619. VAN DER KRAAN, M.I., GROENINK, J., NAZMI, K., VEERMAN, E.C., BOLSCHER, J.G. & NIEUW AMERONGEN, A.V. 2004. Lactoferrampin: a novel antimicrobial peptide in the N1-domain of bovine lactoferrin. Peptides. 25(2): 177-183. VAN DER KRAAN, M.I., NAZMI, K., TEEKEN, A., GROENINK, J., VAN 'T HOF, W., VEERMAN, E.C., BOLSCHER, J.G. & NIEUW AMERONGEN, A.V. 2005a. Lactoferrampin, an antimicrobial peptide of bovine lactoferrin, exerts its candidacidal activity by a cluster of positively charged residues at the C-terminus in combination with a helix-facilitating N-terminal part. Biol Chem. 386(2): 137-142. VAN DER KRAAN, M.I., NAZMI, K., VAN 'T HOF, W., AMERONGEN, A.V., VEERMAN, E.C. & BOLSCHER, J.G. 2006. Distinct bactericidal activities of bovine lactoferrin peptides LFampin 268-284 and LFampin 265-284: Asp-Leu-Ile makes a difference. Biochem Cell Biol. 84(3): 358-362. VAN DER KRAAN, M.I., VAN MARLE, J., NAZMI, K., GROENINK, J., VAN 'T HOF, W., VEERMAN, E.C., BOLSCHER, J.G. & NIEUW AMERONGEN, A.V. 2005b. Ultrastructural effects of antimicrobial peptides from bovine lactoferrin on the membranes of Candida albicans and Escherichia coli. Peptides. 26(9): 1537-1542. WADE, D., BOMAN, A., WAHLIN, B., DRAIN, C.M., ANDREU, D., BOMAN, H.G. & MERRIFIELD, R.B. 1990. All-D amino acid-containing channel-forming antibiotic peptides. Proc Natl Acad Sci U S A. 87(12): 4761-4765. 130 WASSEF, M.K., FIORETTI, T.B. & DWYER, D.M. 1985. Lipid analyses of isolated surface membranes of Leishmania donovani promastigotes. Lipids. 20(2): 108-115. WebPath: The Internet Pathology Laboratory. Special Stains in Histology - Mycobacterium avium complex, Ziehl-Neelsen acid fast stain. Available on: http://library.med.utah.edu/WebPath/HISTHTML/STAINS/STAIN018.html. Access on: 25 August 2011 WHO. Combat Antimicrobial Resistance. Fact Sheet. Available on http://www.who.int/world-health-day/2011/WHD201_FS_EN.pdf. Access on: 27 July 2011. WHO. Leishmaniasis. Available on: http://www.who.int/leishmaniasis/en/. Access on: 27 July 2011. WHO. World Health Day – 7 April 2011. Slides on antimicrobial resistance. Available on: http://www.who.int/world-health-day/2011/WHD_AMR.pps. Access on: 27 July 2011. WHO, E.C.O.T.C.O.T.L. 2010. Control of the leishmaniases : report of a meeting of the WHO Expert Committee on the Control of Leishmaniases, Geneva, 22-26 March 2010, Geneva, World Health Organization. WIESNER, J. & VILCINSKAS, A. 2010. Antimicrobial peptides: The ancient arm of the human immune system. Virulence. 1(5): 440-464. YAMAUCHI, K., TOMITA, M., GIEHL, T.J. & ELLISON III, R.T. 1993. Antibacterial activity of lactoferrin and a pepsin-derived lactoferrin peptide fragment. Infect Immun. 61(2): 719-728. YANG, D., BIRAGYN, A., KWAK, L.W. & OPPENHEIM, J.J. 2002. Mammalian defensins in immunity: more than just microbicidal. Trends Immunol. 23(6): 291-296. YOO, Y.C., WATANABE, R., KOIKE, Y., MITOBE, M., SHIMAZAKI, K., WATANABE, S. & AZUMA, I. 1997a. Apoptosis in human leukemic cells induced by lactoferricin, a bovine milk protein-derived peptide: involvement of reactive oxygen species. Biochem Biophys Res Commun. 237(3): 624-628. 131 YOO, Y.C., WATANABE, S., WATANABE, R., HATA, K., SHIMAZAKI, K. & AZUMA, I. 1997b. Bovine lactoferrin and lactoferricin, a peptide derived from bovine lactoferrin, inhibit tumor metastasis in mice. Jpn J Cancer Res. 88(2): 184-190. ZAIOU, M. 2007. Multifunctional antimicrobial peptides: therapeutic targets in several human diseases. J Mol Med (Berl). 85(4): 317-329. ZUCCA, M. & SAVOIA, D. 2011. Current developments in the therapy of protozoan infections. Open Med Chem J. 5: 4-10. ZWEYTICK, D., TUMER, S., BLONDELLE, S.E. & LOHNER, K. 2008. Membrane curvature stress and antibacterial activity of lactoferricin derivatives. Biochem Biophys Res Commun. 369(2): 395-400.