Studies on the genetics of artemisinin resisteance in malaria

Studies on the genetics of artemisinin resistance in malaria Ana Júlia Pinto Fonseca Sieuve Afonso Rodrigues Universidade Nova de Lisboa Instituto de Higiene e Medicina Tropical Centro de Malária e outras Doenças Tropicais – Laboratório Associado 2007 2 3 Ana Júlia Pinto Fonseca Sieuve Afonso Rodrigues Studies on the genetics of artemisinin resistance in malaria Dissertação de candidatura ao grau de Doutor em Parasitologia Médica submetida ao Instituto de Higiene e Medicina Tropical Universidade Nova de Lisboa Orientadores – Professor Doutor Virgílio Estólio do Rosário Professor Doutor Pedro Vítor Lemos Cravo 4 5 Ana Júlia Pinto Fonseca Sieuve Afonso Rodrigues foi bolseira da Fundação para a Ciência e Tecnologia do Ministério da Ciência Tecnologia e Ensino Superior SFRH/BD/8913/2002 6 7 I declare that unless stated otherwise, the work presented here is my own. A number of figures and tables have been kindly made available by other scientists or developed in partnership with other scientists. Ana Júlia Pinto Fonseca Sieuve Afonso Rodrigues 8 9 Ítaca Se um dia partires rumo a Ítaca, reza para que o caminho seja longo, cheio de aventura e de conhecimento. Não temas monstros como os Ciclopes ou o zangado Poseidon: Nunca os encontrarás no teu caminho enquanto mantiveres o teu espírito elevado, enquanto uma rara excitação agitar o teu espírito e o teu corpo. Nunca encontrarás os Ciclopes ou outros monstros a não ser que os tragas contigo dentro da tua alma, a não ser que a tua alma os crie em frente a ti. Deseja que o caminho seja bem longo para que haja muitas manhãs de Verão em que, com quanto prazer, com tanta alegria, entres em portos que vês pela primeira vez; Para que possas parar em postos de comércio fenícios aí comprar coisas finas, madrepérola, coral e âmbar, e perfumes sensuais de todos os tipos tantos quantos puderes encontrar; e para que possas visitar muitas cidades egípcias e aprender e continuar sempre a aprender com os seus escolares. Tem sempre Ítaca na tua mente. Chegar lá é o teu destino. Mas não te apresses na viagem. Será melhor que ela dure muitos anos para que sejas velho quando chegares à ilha, rico com tudo o que encontraste no caminho, sem esperares que Ítaca te traga riquezas. Ítaca deu-te a tua bela viagem. Sem ela não terias sequer partido. Não tem mais nada a dar-te. E, sábio como te terás tornado, tão cheio de sabedoria e experiência, já terás percebido, à chegada, o que significa uma Ítaca. Konstantinos Kaváfis (1863-1933) . 16 17 TABLE OF CONTENTS 18 19 Page Acknowledgements / Agradecimentos 11 Table of Contents 17 List of Figures 23 List of Tables 27 List of Abbreviations 31 Abstract 35 Resumo 39 Chapter I – Introduction 43 1.1 Malaria: general features 45 1.2 Malaria today 47 1.3 The parasite and its life - cycle 50 1.4 The genetics of malaria parasites 53 1.5 Antimalarial drugs and targets 57 1.6 Artemisinin and its derivatives 58 1.6.1 Artemisinin mode of action 63 1.7 Drug Resistance 67 1.7.1 Resistance to artemisinin 71 1.8 Rodent malaria parasites 7 3 1.8 .1 The geneti cs of rodent malaria parasites 78 1.9 Linkage Group Selection 80 1.9.1. Amplified Fragment Length Polymorphism 83 1.10 Aims of the project 8 4 Chapter II – Materials and Methods 8 5 2.1. Mice 87 2.2. Mosquitoes 87 2.3. Parasites 87 2.4 Preparation of 10 7 standard parasite inocula 88 2.5 Cloning 88 2.6 Preparation and administration of artemisinin and artesunate 88 2.7 Drug tests 89 2.8 Artemisinin and artesunate selection experiments: general procedure 89 2.9 Tests to evaluate the stability of drug-resistance 90 2.10. Transmission through Anopheles stephensi 90 2.11. DNA extraction 91 2.12. Identification of the P. chabaudi tctp and atp6 genes 9 1 2.13. Amplification and sequencing of the mdr1, cg10, tctp and atp6 genes of P. chabaudi 92 2.14. Estimation of copy numbers of the pcmdr1, pctctp and pcatp6 genes 92 2.15 Production of cross progeny - Overview of procedure 93 2.15.1 Mosquito feeds 9 4 2.15.2 Infection of mice from sporozoites 95 2.15.3 Preparation of artemisinin and artesunate 95 2.16 Linkage Group Selection 9 5 2.16.1 Selection of cross progeny 9 5 2.16.2 Amplified Fragment Length Polymorphism (AFLP) analysis 96 2.16.2.1 Preparation of parasite DNA from experimental groups 96 2.16.3 AFLP analysis 9 7 2.16.4 Measurement and c omparison of the intensity of AFLP markers 10 1 2.16.5 Assignment of AFLP markers to locations in a P. chabaudi genetic linkage map. 102 2.16.6 Sequencing of AFLP markers, and their location on the Plasmodium falciparum genome 103 2.16.7 AFLP band sequen cing and purification of PCR products 10 3 2.17 Experiments with genetic crosses of AS-ART and AJ or AS-ATN and AJ 105 2.17.1 Blood collection for Proportional sequencing analyses 105 2.17.2 Extraction of parasite DNA from blood or sporozoites 10 6 2.17. 3 Determination of the proportions of clones in genetic crosses by proportional sequencing 10 6 2.17.3.1 Principle of proportional sequencing 106 2.17.3.2 Proportional sequencing - PCR reagents and reaction conditions 10 9 20 2.17.3.3 Proportional sequencing – Sequencing and purification of PCR products 1 10 2.17.3.4 Proportional sequencing – Analyses of sequencing results 110 2.18. Amplification and sequencing of the ubp-1 gene of P. chabaudi 112 Results 11 3 Chapter III – Experiments for drug selection of artemisinin and artesunate resistance 11 7 3.1 Introduction 119 3.2 Artemisinin drug selection 120 3. 3 Artesunate drug selection 12 4 3. 4 Cloning of resistant parasites 1 2 8 3.5 Drug resistance stability tests 130 3.5.1. Resistance stability after liquid nitrogen preservation (deep-freezing) 130 3.5.2. Resistance stability after blood passages in the absence of drug pressure 131 3. 5 .3. Resistance stability after cyclical transmission through mosquitoes 1 31 3.6 Test for cross-resistance between artemisinin and artesunate clones 133 3.7 Discussion 134 Chapter IV – Analysis of the putative genetic modulators for artemisinin and artesunate resistance 137 4. 1 . Sequencing of P. chabaudi mdr1 , cg10, tctp and atp6 genes, in the selected mutant clones. 1 3 9 4.1.1. Isolation of the mdr1, cg10, tctp and atp6 P.chabaudi orthologues 139 4.2.2. Sequnece comparions of P. chabaudi mdr1, cg10, tctp and atp6 in the selected mutant clones and their sensitive progenitors. 143 4.2. Estimation of gene copy numbers of the pcmdr1, pctctp and pcatp6 genes, in the selected mutant clones. 144 4.3. Discussion 146 4.3.1. Summary 14 6 4.3.2. General discussion 147 Chapter V – Experiments using Linkage Group Selection as an attempt to identify the locus or loci involved in artemisinin and artesunate resistance 149 5.1. Introduction 151 5.2 Production of cross progeny 1 52 5.3 Selection of cross progeny 1 53 5.3.1 General procedure 153 5.3.2 Results 154 5.3.2.1 Selection of the AS - ART x AJ cross 1 55 5.3.2.2 Selection of the AS - ATN x AJ cross 1 58 5.4 AFLP analysis 160 5.4.1 General procedure 160 5.4. 2 Results 1 61 5.4.2.1 AFLP analysis for the AS - ART x AJ 1 62 5.4.2.2 AFLP analysis for the AS-ATN x AJ 165 5.4.2.3 AFLP analysis comparison between pooled crosses AS-ART x AJ and AS-ATN x AJ 169 5.5 Sequencing and mapping of AFLP bands under selection 17 0 5.5.1 General procedure 170 5.5.2 Results 171 5.6 Further investigation of markers under selection on chromosome 2 1 73 5.6.1 Identification of ubp - 1 gene mutation o n P. chabaudi chromosome 2 1 75 5.7 Discussion 178 Chapter VI – General conclusions 181 Supplement 185 Appendix 1 - Solutions and buffers 1 87 Appendix 2 – Gene and primer sequence for the genes pcmdr1, cg10, pctctp and pcatp6 189 Appendix 3 – Primer sequence and PCR amplification conditions for determining gene copy number of pcmdr1, pctctp and pcatp6 genes 193 Appendix 4 – Gene and primer sequence for the gene ubp-1 194 Appendix 5 – The relative intensity and comparative intensities of all AFLP markers analysed in the LGS experiments described in Chapter V 202 21 Glossary 2 21 References 2 25 Online references 2 39 Publications that arose from this project 243 Hunt P, Afonso A , Creasey A, Culleton R, Sidhu A, Logan J, Valderramos S, McNae I, Cheesman S, do Rosario V, Carter R, Fidock D and Cravo P. Gene encoding a de-ubiquitinating enzyme is mutated in artemisininand chloroquine-resistant rodent malaria parasites. In press at the Molecular Microbiology Cheesman S, Creasey A, Degnan K, Kooij T, Afonso A , Cravo P, Carter R, Hunt P. Validation of Pyrosequencing for accurate and high throughput estimation of allele frequencies in malaria parasites Mol Biochem Parasitol. 2007 Apr; 152 (2): 213-219. Cravo P, Culleton R, Afonso A , Ferreira ID, do Rosário VE Mechanisms of Drug Resistance in Malaria: Current and New Challenges. Anti-Infective Agents in Medical Chemistry 2006. 5: 63-73. Afonso A , Hunt P, Cheesman S, Alves AC, Cunha CV, do Rosário V, Cravo P Malaria parasites can develop stable resistance to artemisinin but lack mutations in candidate genes atp6 (serca), tctp, mdr1 and cg10. AntiMicrobial Agents Chemotherapy 2006. 50 (2): 480-489. 22 23 LIST OF FIGURES 24 25 Page Cover - Artemisia annua flowers and leaves. Source: The University of York (www.york.ac.uk) Figure 1 - Illustration drawn by Laveran of various stages of malaria parasites as seen on fresh blood. 46 Figure 2 – World geographic distribution of malaria, data from 2003. 49 Figure 3 – Malaria tra nsmission areas and P. falciparum drug resistance distribution data from World Heath Organization data from 2004 49 Figure 4 - Life cycle of malaria parasite. 52 Figure 5 – Crossing and chromosomal events in Plasmodium. 55 Figure 6 - Schematic represent ation of Artemisia annua L. 58 Figure 7 – A - Chemical structure of (1) Qinghaosu or artemisinin and some of its derivatives (2) dihydroartemisinin, (3) artemether; (4) arteether and (5) artesunic acid or artesunate. 60 Figure 8 The geographic origins of the rodent malaria parasites. 7 3 Figure 9 Plasmodium chabaudi parasites (trophozoite stage) in mouse peripheral blood. 7 5 Figure 10 – Rodent malaria genomes synteny map to P. falciparum. 79 Figure 11 – Schematic representation of the Linkage Group Selection protocol. 82 Figure 12 - An example of an AFLP gel. 10 0 Figure 13 - The four polymorphisms on the pcdh ps gene that allowed differentiating between strains AS and AJ. 10 8 Figure 14 – Example of an electropherogram for proportional sequence analysis of two polymorphisms on the pcdhps gene. 111 Figure 15 – A schematic representation of the artemisinin selection procedure. AS - ART* is uncloned. 1 20 Figure 16 – The increase of the artemisinin dose (mg/kg/day), tolerated by AS-30CQ clone, during 15 blood passages. 121 Figure 17 – The mean day of recrudescence on each passage under selection and artemisinin dose (mg/kg/day). 12 2 Figure18 – A schematic representation of the artesunate selection procedure. AS-ATN* is uncloned. 124 Figure 19 – The increase of the artesunate dose (mg/kg/day), tolerated by AS-15CQ clone, during 14 blood passages. 125 Figure 20 – The mean day of recrudescence under selection and artesunate dose (mg/kg/day). 1 2 6 Figure 21 – A schematic representation of the clones and parasite lines of Plasmodium chabaudi used in this project. 128 Figure 22 – A schematic representation of the artemisinin and artesunate selection procedure. 13 2 Figure 23 – Extent of sequence analysed for the gene pcmdr1 for the clones AS - 15CQ, AS - 30CQ, AS-ART and AS-ATN. 1 41 Figure 24 – Extent of sequence analysed for the gene pccg10 for the clones AS-15CQ, AS-30CQ, ASART and AS-ATN. 141 Figure 2 5 – Extent of sequence analysed for the gene p ctctp for the clones AS - 15CQ, AS - 30CQ, AS - ART and AS-ATN. 14 2 Figure 2 6 – Extent of sequence analysed for the gene pcatp6 for the clones AS-15CQ, AS-30CQ, ASART and AS-ATN 142 Figure 27 – Relative differences (N - fold) in gene copy number between artemisinin (AS - ART) and artesunate (AS-ATN) resistant parasites and their sensitive progenitors, AS (30CQ) and AS (15CQ) respectively. 1 4 5 Figure 28 - Schematic representation of the selection of the pooled cross progeny (LGS) experiment using previously generated individual crosses between AJ and AS-ART 1 53 Figure 2 9 - Parasitaemia curves for the “untreated” and “artemisinin treated” uncloned cross progeny of the AS-ART x AJ cross. 155 Figure 30 – Detail of the pcdhps gene sequence with indication of polymorphisms 1 - 4 between AS and AJ. For entire sequence of the gene please check figure 13. 1 56 Figure 31 - Parasitaemia curves for the “untreated” and “artesunate treated” uncloned cross progeny of the AS-ATN x AJ cross. 158 Figure 32 - Comparative Intensities (CI) of AJ (sensitive parent) specific AFLP markers in the ART treated group compared to the untreated group, for the AS-ART x AJ cross. 1 63 Figure 33 - Comparative Intensities (CI) of AS (resistant parent) specific AFLP markers in the ART treated group compared to the untreated group, for the AS-ART x AJ cross 164 Figure 34 - Comparative Intensities (CI) of AJ (sensitive parent) specific AFLP markers in the ATN treated group compared to the untreated group, for the AS-ATN x AJ cross. 1 67 Figure 35 - Comparative Intensities (CI) of AS (resistant parent) specific AFLP markers in the ATN treated group compared to the untreated group, for the AS-ATN x AJ cross. 168 Figure 36 - P. chabaudi chromosome 2 with syntenic blocks represented, P. falciparum chromosome 1 74 32 33 3D7 P. falciparum culture, sensitive 6-PGD - 6-Phosphogluconate dehydrogenase A - Adenine (in DNA context) A Alanine (Ala) (in protein context) a.a. - Amino acid ACT - Artemisinin combination therapy AFLP - Amplified fragment length polymorphism ART - Artemisin in AJ - Plasmodium chabaudi clone, totally drug sensitive. Genetically different from AS line . AS-ART - Plasmodium chabaudi clone, pyrimethamine-, high chloroquineresistant, obtained from AS-30CQ by artemisinin drug pressure (clone obtained during this project) AS-ART* - Plasmodium chabaudi parasite line, pyrimethamine - , high chloroquine - resistant, obtained from AS-30CQ by artemisinin drug pressure (parasite line obtained during this project), this parasite line was then cloned by limiting dilution AS-ATN - Plasmodium chabaudi clone, pyrimethamine-, intermediate chloroquineresistant, obtained from AS-15CQ by artesunate drug pressure (clone obtained during this project) AS-ATN* - Plasmodium chabaudi parasite line, pyrimethamine - , intermediate c hloroquine - resistant, obtained from AS-15CQ by artesunate drug pressure (parasite line obtained during this project), this parasite line was then cloned by limiting dilution AS-SENS - Plasmodium chabaudi clone, totally drug sensitive. Genetically differ ent from AJ. AS-PYR - Plasmodium chabaudi clone, pyrimethamine - resistant obtained from AS - SENS by pyrimethamine drug pressure AS-3CQ - Plasmodium chabaudi clone, pyrimethamine and low chloroquine resistant obtained from AS-PYR by chloroquine drug pressure AS-15CQ - Plasmodium chabaudi parasite isolate, pyrimethamine and intermediate chloroquine resistant, obtained from the AS-3CQ by cloroquine drug pressure AS-15MEF - Plasmodium chabaudi parasite clone, pyrimethamine, intermediate chloroquine resistant and mefloquine resistant, obtained from the AS-15CQ by mefloquine drug pressure AS-30CQ - Plasmodium chabaudi parasite clone, pyrimethamine and high chloroquine resistant, obtained from the AS-15CQ by cloroquine drug pressure ARMD - Accelerated resistant to multiple drugs phenotype ama-1 - Apical m embrane a ntigen - 1 ATN - Artesunate atp6 - Encoding the Sarcoplasmic and Endoplasmic Reticulum Ca 2+ atpase gene C - Cysteine (Cys) CDC - National Center for Disease Control cDNA - Complementary DNA cg10 - Gene coding for a putative protein transporter, P. chabaudi orthologue of the P. falciparum crt. CI - Comparative i ntensity CQ - Chloroquine crt - Chloroquine resistance transporter gene CSP - Circumsporozoite protein D - Aspartic acid (Asp) Dd2 P. falciparum culture, from Indochina, pyrimethamine , mefloquine and chloroquine resistant DDT - Dichloro-diphenyl-trichloroethane DHA - Dihydroartemisinin DHFR Dihydrofolate - reductase enzyme DMSO - Dimethyl sulfoxide DNA - Deoxyribonucleic acid F - Phen ylalanine (Phe) gDNA - Genomic DNA HB3 - P. falciparum culture, from Honduras, pyrimethamine, mefloquine and chloroquine resistant IC 50 - The drug dose necessary to eliminate 50% of the parasites II Intensity index i. p. - Intraperitoneally iRBC - Infected red blood cells (erythrocytes) K i Biochemistry catalytic constant 34 L Leucine (Leu) LDH - Lactate dehydrogenase LGS - Linkage group selection Mb - Mega base MCD - Minimum c urative d ose mdr1 - Multi d rug r esistance 1 gene msp-1 Merozoite surface protein-1 msp-2 - Merozoite surface protein-2 N - Asparagine (Asn) NADH - Hydrogen n icotinamide a denine d inucleotide NCBI/NIH - National Institute of Health nM - Nanomolar PABA - Paraminobenzoic a cid pcdhps - Gene of Plasmodium chabaudi codifying for enzyme dihydropteroate synthetase PCR - Polymerase chain reaction Phe - Phenylalanine PYR - Pyrimethamine RBC - Red blood c ells (erythrocytes) RIIs - Relative intensity indices RMP - Rodent malaria parasite RNA - Ribonucleic a cid rRNA - Ribosomal RNA RTQ - PCR - Real time quantitative – PCR S - Serine (Ser) s. c . - Subcutaneous SDS - Sodium dodecyl sulfate sRNA - Small RNA SP - Sulfadoxine-pyrimethamine T - Tymine tctp - Translationally controlled tumour protein gene tRNA - Transfer RNA Tyr - Tyrosine ubp-1 - De - ubiquitinating enzyme, u biquitin carboxyl - terminal hydrolase, putative, 1 ubp-1 De - ubiquitinating enzyme, u biquitin carboxyl - terminal hydrolase, putative, 1 gene V - Valine (Val) WHO - World Heath Organization Y - Tyrosine (Tyr) 6-PGD - 6-Phosphogluconate dehydrogenase 35 ABSTRACT 36 37 Resistance of Plasmodium falciparum to multiple drugs including chloroquine (CQ) and sulfadoxine-pyrimethamine (SP) is a major problem in malaria control. New drugs, such as artemisinin (ART) derivatives, particularly in combination with other drugs, are thus increasingly used to treat malaria. Although stable resistance to ART has yet to be reported from laboratory or field studies, its emergence would be disastrous because of the lack of alternative treatments. The work presented in this thesis describes the selection of parasites with stable resistance to ART and artesunate (ATN), and their genetic analysis. This work was carried out using the rodent malaria parasite Plasmodium chabaudi chabaudi (Plasmodium chabaudi). Two different rodent malaria parasite lines AS-15CQ and AS-30CQ were continually passaged in the presence of increasing concentrations of ATN or ART, respectively. After selection, these lines, named AS-ATN and AS-ART, showed 6-fold and 15-fold increased resistance to ATN and ART respectively. Resistance remained stable after cloning, freeze/thawing, blood passage in the absence of drug pressure and transmission through mosquitoes. The nucleotide sequences and the gene copy number of the possible genetic modulators of ART resistance mdr1, cg10, tctp and atp6; were compared between sensitive and resistant parasites. No mutations or changes in the gene copy number of these genes were found. Linkage Group Selection (LGS) was used to investigate the genetic basis of ART resistance. Genetic crosses between AS-ART or AS-ATN and the ART-sensitive clone AJ were analysed before and after drug treatment. Using quantitative markers, a genetic locus on chromosome 2 was found to be under strong selection. Loci on chromosomes 1, 8 and 14 of P. chabaudi also appear to be under selection. On chromosome 2, two different mutations V739F and V770F in a de-ubiquitinating enzyme (ubp-1) were identified in AS-ATN and AS-ART respectively, relative to their sensitive progenitors. The implications of these results are discussed. 38 39 RESUMO 40 41 A existência de estirpes do parasita, Plasmodium falciparum resistentes a multiplos fármacos tais como; cloroquina (CQ) e sulfadoxina-pirimetamina (SP) é um dos problemas mais graves no controlo da malária. Novos fármacos, como a artemisinina (ART) e seus derivados, particularmente em combinação com outros fármacos, são cada vez mais utilizados no tratamento da malaria. Embora até ao momento a fármaco-resistência estável à ART quer in vitro quer in vivo não tenha sido registada, o seu surgimento seria desastroso devido á falta de alternativas. O trabalho apresentado nesta tese descreve a selecção de resistência estável a ART e ao artesunato (ATN). Este trabalho foi realizado usando o modelo roedor de malária Plasmodium chabaudi chabaudi (Plasmodium chabaudi). Duas linhas parasitáricas diferentes, AS-15CQ e AS-30CQ, foram feitas crescer na presença de concentrações crescentes de ATN e ART, e que no final apresentavam uma resistência de 6 e 15 vezes superior ao ATN e à ART, respectivamente (estas novas linhas obtidas foram nomeadas AS-ATN e AS-ART). A resistência é estável mesmo após clonagem, congelamento/descongelamento, passagem sanguínea na ausência de pressão de fármaco e transmissão através do mosquito vector. A sequência nucleotídica e o número de cópias dos genes descritos como moduladores putativos de resistência à ART: mdr1, cg10, tctp e atp6; foi comparada entre parasitas resistentes e sensíveis. Não tendo sido encontradas alterações na sequência ou no número de cópias destes genes. Numa tentativa de identificar os genes encolvidos na resistância à ART e ao ATN a técnica de Linkage Group Selection (LGS) foi utilizada e dois cruzamentos genéticos entre os clones fármaco-resistentes; AS-ART e AS-ATN e o clone geneticamente distinto dos anteriores e sensível aos fármacos em estudos; AJ; foram realizados. Foram encontrados sobre selecção em ambos os cruzamentos genéticos quatro loci; cromossomas de P. chabaudi 1, 2, 6 e 8. Atendendo a que, a selecção no cromossoma 2 era mais forte, este locus foi submetido a análises genéticas subsequentes. Tendo sido encontradas duas mutações diferentes (V739F e V770F) num gene que codifica para um enzima de desubiquitinação (ubp-1). As implicações destes resultados serão discutidas. 48 P. falciparum has become resistant to almost all drug classes except the artemisinin derivatives. Nowadays chloroquine-resistant P. falciparum occurs across all malaria endemic areas. The effectiveness of sulfadoxine-pyrimethamine has rapidly declined in all regions where it has been introduced due to resistance, and multidrug resistance is now established in Southeast Asia, South America and Africa (See Figure 3) [Collins WJ et al. 2006; Green MD 2006; Kshirsagar NA 2006; Linares GE et al. 2007]. Drug resistance is most likely to emerge when background immunity is weak, parasite numbers in an individual are high, transmission is low and drug pressure is intense or very intense [Hastings IM et al. 2000]. With an increase in insecticide and antimalarial-drug resistance, the development of a malaria vaccine and above all new drugs or new drug combinations, using drugs already in use, carries huge expectations [Chatterjee S et al. 2006; Girard MP et al. 2007; Greenwood B et al. 2007; Matuschewski K 2006, Shanks GD 2006; Stepniewska K et al. 2006]. 49 Figure 2 – World geographic distribution of malaria, data from 2003. Source: National Center for Disease Control (CDC) with kind permission of CDC. Figure 3 – Malaria transmission areas and P. falciparum drug resistance distribution data from World Heath Organization data from 2004. Source: World Heath Organization (WHO) with kind permission of WHO. 50 1.3 The parasite and its life-cycle In nature, malaria parasites spread by infecting successively two types of hosts: humans and female Anopheles mosquitoes. Malaria is transmitted among humans by female mosquitoes of the genus Anopheles. Female mosquitoes require blood meals in order to carry out egg production, and such blood meals are the link between the human and the mosquito hosts in the parasite life cycle. Of the approximately 430 known species of Anopheles, only 30-50 transmit malaria in nature (“vectors”). The successful development of the malaria parasite in the mosquito (from the “gametocyte” stage to the “sporozoite” stage – See Figure 4) depends on several factors. The most important is ambient temperature and humidity and whether the Anopheles survives long enough to allow the parasite to complete its cycle in the mosquito host (“sporogonic” or “extrinsic” cycle, duration 10 to 18 days). Malaria’s life cycle is comprised of both the sexual and asexual forms (See Figure 4). The sexual cycle occurs mainly in the mosquito; while the asexual cycle takes place in the human host after the parasites have entered the host’s blood stream when the mosquito bites for a blood meal. During a blood meal, a malaria infected female Anopheles mosquito inoculates sporozoites into the human host. Though the salivary glands of an infected mosquito contain thousands of sporozoites, less than 100 of these are transmitted in any one bite [Rosenberg R et al. 1990]. Within 30-45 minutes of the parasite’s sporozoites entering the bloodstream, they enter parenchymal cells of the liver; this is achieved by the binding of the thrombospondin domains of the circumsporozoite and thrombospondin-related adhesive proteins (csp and trap respectively) to the heparin sulphate proteoglygan on the hepatocytes [Frevert U et al. 1993]. This phase is called the preerythrocytic stage lasting 5-15 days in which the parasite undergoes asexual reproduction (schizogony): the end products of this are the merozoites. In P. vivax and P. ovale a dormant stage called hypnozoites, can persist in the liver and can cause relapses by invading erythrocytes weeks or years later [Durante Mangoni E et al. 2003]. Hepatocytes rupture to release merozoites that enter red blood cells. Invasion of erythrocytes by merozoites involve an initial low affinity interaction between proteins on the surface coat of the merozoite (merozoite surface protein-1 (msp-1), and apical membrane antigen-1 (ama1)) and the surface of the erythrocyte. Once inside the erythrocyte the merozoite initiates the feeding process forming the intracellular parasite, the trophozoite (erythrocytic schizogony stage). Mitotic divisions (asexual reproductive stage) occur in the cells giving rise to schizonts, which contain up to about thirty haploid merozoites. This red cell ruptures, releasing more mature merozoites, which invade more red blood cells, hence maintaining this 51 “asexual cycle”. Some parasites undergo gametocytogenesis within the erythrocyte, producing male or female micro and macrogametocytes respectively. These remain in the blood circulation where they are available for ingestion by a feeding mosquito. The asexual reproductive stage, occurring in the blood of the vertebrate host (human, primate or rodent), is the target of most antimalarial drugs, including artemisinin and its derivatives. Inside the mosquito mid-gut, female and male gametocytes undergo gametogenesis, in which the female macrogametocyte escapes from the erythrocyte membrane and the male microgametocyte undergoes the process of exflagellation which produces 8 motile microgametes. The micro and macro-gametes fuse to form a zygote that in turn becomes an ookinete, (the only diploid stage of the parasite). The ookinete crosses the gut wall and encysts on the outer wall of the gut beneath the basal lamella forming an oocyst or sporocyst. Division and multiplication of the sporocyst takes place to produce many haploid sporozoites. When the sporocyst bursts the sporozoites are released and then migrate to the salivary gland, waiting to re-infect again once the mosquito takes another blood-meal [Barnwell JW et al. 1998; Beier JC et al. 1998; Sinden RE 1997]. See Figure 4 for details on the parasite life cycle. 52 Figure 4 - Life cycle of malaria parasite. Most of the biological work presented here occurs in the erythrocytic cycle of the parasite (presented in the Figure as B). The drug resistance selection process occurs on this part of the parasite life cycle. Briefly: the malaria parasite life cycle involves two hosts. During a blood meal, a malaria-infected female Anopheles mosquito inoculates sporozoites into the human host . Sporozoites infect liver cells and mature into schizonts , which rupture and release merozoites . After this initial replication in the liver (exoerythrocytic schizogony ), the parasites undergo asexual multiplication in the erythrocytes (erythrocytic schizogony ). Merozoites infect red blood cells . The ring stage trophozoites mature into schizonts, which rupture releasing merozoites . Some parasites differentiate into sexual erythrocytic stages (gametocytes) . Blood stage parasites are responsible for the clinical manifestations of the disease. The gametocytes, male (microgametocytes) and female (macrogametocytes), are ingested by an Anopheles mosquito during a blood meal . The parasites’ multiplication in the mosquito is known as the sporogonic cycle . While in the mosquito's stomach, the microgametes penetrate the macrogametes generating zygotes . The zygotes in turn become motile and elongated (ookinetes) which invade the midgut wall of the mosquito where they develop into oocysts . The oocysts grow, rupture, and release sporozoites , which make their way to the mosquito's salivary glands. Inoculation of the sporozoites into a new human host perpetuates the malaria life cycle. Source: National Centre for Disease Control (CDC) with kind permission of CDC. 53 1.4 The genetics of malaria parasites Malaria parasites, as all members of the phylum Apicomplexa, are haploid for almost their entire life cycle (exo-erythrocytic and erythrocytic blood stages, sporogony and microgametogenesis), and in the haploid phase of the parasite life cycle they multiply by mitosis. The only phase of the parasite life cycle where the parasite genome is diploid is the zygote stage (ookinetes), prior to the meiotic division that results in the production of sporozoites. Malaria parasites have three individual genomes; an extra-chromosomal mitochondrial genome, a 35kb circular genome and a large nuclear genome. The mitochondrial genome, also known as the 6kb element contains genes encoding two truncated ribosomal sRNA and three proteins components involved in the electron transport system; cytochrome c oxidase subunits I and II and cytochrome b [Funes S et al 2004]. The inheritance of the 6 kb element appears to follow the same pattern as other mitochondrial genomes in eukaryotes meaning it is inherited from the female parent only [Creasey AM et al. 1993]. The 35kb circular genome associated with the apicoplast encodes 30 proteins, mainly rRNA, tRNA, which are primarily involved in gene expression [Funes S et al. 2004, Gardner MJ et al. 2002]. The exact role of the apicoplast remains unclear, but it is known to be involved in the anabolic synthesis of fatty acids, isoprenoids and haem [Gardner MJ et al. 2002]. The haploid nuclear genome of P. falciparum is where most parasite genes reside. It consists of 14 chromosomes and encodes approximately 5,300 genes with a total genome size of 22.8 Mb. The parasite chromosomes have a central domain that contain conserved coding regions and chromosome ends that consist of telomeric repeat sequences and subtelomeric repeat regions, containing polymorphic gene families (for example pfemp1, stevors and rifins) [Lanzer M et al. 1994]. The P. falciparum nuclear genome is very (A+T)-rich, with an overall (A+T) content of 81%, rising to 90% in intronic and intergenic regions [Gardner MJ et al. 2002]. There is considerable chromosomal size polymorphism between strains of parasites [Corcoran LM et al. 1986], which could be due to unequal crossing-over of homologous chromosomes during meiosis, or non-meiotic chromosome breaking and healing events [Babiker HA et al. 1994, Gardner MJ et al. 2002, Hernandez-Rivas R et al. 1996, Scherf A et al. 1992]. Various genetic polymorphisms can be observed when comparing different strains even within the same Plasmodium species. This diversity observed to the genotype level has its origin on either spontaneous genetic mutation, occurring at any stage of the parasite 54 development and also through genetic recombination occurring in the mosquito vector stage. Genetic recombination which the parasites undergo in the mosquito midgut can then result in independent assortment of genes on different chromosomes (See Figure 5) [Walliker D et al. 1998]. When a mosquito feeds on an infected host containing two genetically distinct parasites, gametocytes from the two parasites are taken up, and may fertilise, recombining into heterozygous zygotes, by meiosis, producing four genetically distinct haploid daughter cells, which are called the recombinant progeny (See Figure 5). In the case where there are equal numbers of male and female gametes present from each parental strain in the mosquito, then selfing will occur 50% of the time, resulting in 25% of the zygotes being genetically identical to one parental strain, and 25% identical to the other. The remaining 50% will be hybrid between the two parentals.What is meant is that if the recombinant progeny was undergoing a normal Mendelian segregation pattern meaning without any kind of selfing/crossing bias, thus in the presence of a random segregation and if there are equal numbers of male and female gametes present from each parental strain in the mosquito, then selfing (recombination within the same strain) will occur 50% of the time, resulting in 25% of the zygotes being genetically identical to one parental strain, and 25% identical to the other. The remaining 50% will be hybrid between the two parentals. 55 A B C D Figure 5 – Crossing and chromosomal events in Plasmodium. Clone 1 and clone 2 are gametes (haploid stage) from two genetically distinct parasites, after the blood meal zygotes are formed. The formation of these zygotes can result from selfing (equal to the progenitor gametes) or from crossing (cross between different clones and are therefore heterozygous). Through meiosis four genetically distinct haploid daughter cells are produced, which are called the recombinant progeny. A and D are the result of selfing on the other hand B and C are the product of recombination. From: Walliker D 2000, with kind permission of Professor David Walliker. Recombination between malaria parasites was first proven with genetic crossing experiments done with the rodent parasites Plasmodium yoelii [Walliker D et al. 1971], and Plasmodium chabaudi [Walliker D et al. 1975]. In the P. chabaudi experiments, two cloned parasites 56 which differed in their response to the anti-malarial drug pyrimethamine and in the electrophoretic patterns of two enzymes (6-phosphogluconate dehydrogenase (6-PGD) and lactate dehydrogenase (LDH)) were mixed in mosquitoes and the resulting progeny were cloned and characterized for their enzyme type and their phenotypic response to pyrimethamine. It was found that not only had the two enzyme isoforms recombined, but that pyrimethamine susceptibility segregated independently [Walliker D et al. 1975], which showed that recombination between the parental characters had occurred. In P. falciparum the production of heterozygotes (in the oocyst) between two heterologous malaria parasites has also been demonstrated experimentally, by dissecting individual oocysts from mosquitoes that had fed on a mixed P. falciparum blood infection of clones 3D7 and HB3. After performing genetic typing of alleles of msp-1 and msp-2 genes, it was found that some oocysts contained alleles exclusively from HB3, some contained alleles only from 3D7, and the remainder of the oocysts contained alleles from both parents, and were therefore hybrids, meaning, the products of fertilization between the two different parental strains. The proportion of the homozygous and heterozygous forms was consistent with random fertilization between parents [Ranford-Cartwright L et al. 1993]. With the objective of sequencing the genome of the human malaria parasite Plasmodium falciparum (clone 3D7), an International Malaria Genome Sequencing Consortium was formed in 1996. The genome was sequenced by three groups: The Institute for Genomic Research and the Malaria Program of the Naval Medical Research Center (chromosomes 2, 10, 11 and 14), The Wellcome Trust Sanger Institute (chromosomes 1, 3-9, 13) and Stanford University (chromosome 12). In 2002, the complete genome of Plasmodium falciparum was published, triggering the “postgenomic” age of malariology [Gardner MJ et al. 2002]. 57 1.5 Antimalarial drugs and targets Antimalarial drugs are one of the most important measures to control the disease. The drug of choice depends on the parasite species and local conditions, drug resistance prevalence and specificity. Traditionally, antimalarial agents are classified as blood schizontocides, tissue schizonticides, gametocides and sporontocides, depending on the stages of the malaria life cycle which are targeted by the drug [Tracey J et al. 1996]. For details see Table 1. Blood schizontocides are drugs acting on asexual intraerythrocytic stages of malarial parasites. They suppress the proliferation of plasmodia in the erythrocytes. Tissue schizontocides prevent the development of hepatic schizonts. They are causally prophylactic because they affect the early developmental stages of the protozoa and prevent the invasion of the erythrocytes. A hypnozoiticide acts on persistent intrahepatic stages of P. vivax and P. ovale in the liver. Gametocides destroy the intraerythrocytic sexual forms (gametes) of the protozoa and the prevent transmission from human to another mosquito. Antimalarials are rarely used clinically just for their gametocidal action Table 1 – Classification of antimalarial agents according to their stage of action. Stage of Action Antimalarial Tissue schizontocides Primaquine, pyrimethamine, sulfonamides (and other 8aminoquinolines and other folate inhibitors) Hypnozoiticides Primaquine, tafenoquine Blood schizontocides Type 1, quick onset : Chloroquine, mefloquine, quinine, halofantrine, artemisinin Type 2, slow onset: Pyrimethamine, sulfonamides, sulfones, other antibiotics, atovaquone Gametocides Primaquine for P. falciparum Quinine for P. vivax, P. malariae and P. ovale Sporontocides Primaquine, chloroquine 64 have also shown that the breakdown of artemisinin results in free radicals [Meshnick SR et al 1996] during this artemisinin breakdown process ferryl ions (Fe[IV]=O) appear to be formed [Kapetanaki S et al 2000]. There are other electrochemical studies that have shown that heme/iron can catalyse the irreversible breakdown of artemisinin derivatives [Zhang F et al.1992] and also structure–activity relationship studies have shown a high correlation between antimalarial activities and heme-binding [Meshnick SR et al 1996] and between antimalarial activity and protein alkylation [Meshnick SR et al 1996]. Also, the predicted pharmacophore or drug receptor from several structure–activity relationship studies seems to resemble heme [Meshnick SR et al 1996] and also there are a lot of other theoretical studies that have shown that artemisinin could bind to and react with heme itself [Gu HM et al. 1984] furthermore, the presence of the heme polymer, hemozoin, is associated with sensitivity to artemisinins for example artemisinins are inactive against the RC strain of Plasmodium berghei [Peters W et al. 1986], and the related intraerthrocytic apicomplexan parasite, Babesia microti [Wittner M et al. 1996],.which both lack hemozoin, yet are most active against schistosomes which also produce hemozoin [Utzinger J et al.2001]. Though all these evidences one could not say that artemisinins act like the typical oxidating drugs which cause promiscuous damage to protein, nucleic acids and lipid firstly because, unlike most other oxidant drugs (and oxidizing agents per si), artemisinin cannot be cyclically oxidised and reduced causing the cascade effect typical from free radical reactions [Zhang F et al.1992] in a way that only one free radical can result from one drug molecule, secondly, all of the oxidant end products observed experimentally were only observed at very high drug concentrations [Berman PA et al. 1997], but the drug is effective at much lower concentrations. This is a very strong indication that, artemisinin derivatives must have a more selective toxic effect. One suggestion to the selective toxicity of artemisinins may be the formation of covalent adducts with parasite components, which will then serve as mediators for free radical intermediates. One important alkylation target is heme itself. Artemisinin– heme adducts have been demonstrated in parasite cultures treated with therapeutic concentrations of artemisinin derivatives [Hong YL et al. 1994] this means that heme is both an activator and target of the artemisinin derivatives. The modification that occurs in heme via its ligation to artemisinin could kill the parasite in several ways, firstly, artemisinin or its heme adduct might be able to inhibit hemozoin biosynthesis or cause hemozoin degradation, for example Pandev and co-workers proved that at micromolar concentrations, artemisinin inhibits hemoglobin digestion by malaria parasites and inhibits hemozoin formation [Pandey AV et al. 1999] though this observation has only been demonstrated in cell-free conditions, artemisinin treatment of living intraerythocytic P. 65 falciparum in culture caused no change in hemozoin content [Asawamahasakda W et al. 1994], suggesting that heme metabolism might not be the major intracellular target. But heme is not the only protein that artemisinin forms adducts with, artemisinins also form covalent adducts with other protein, but not with DNA [Yang YZ et al 1993, Yang YZ et al. 1994]. The alkylation of specific malaria proteins by artemisinins has been demonstrated [Asawamahasakda W et al. 1994] and this could mediate the killing action of artemisinin derivatives since it occurred at therapeutic concentrations of drug. One of the major alkylation targets described in the literature is the malarial translationally controlled tumour protein (tctp protein), a protein that binds heme [Bhisutthibhan J et al. 1998], also previous microscopic studies had proven that some of the malarial tctp protein is present in the food vacuole membranes, where it is in proximity to the heme-rich food vacuole [Bhisutthibhan J et al. 1998]. Thus, it is likely that the reaction between artemisinin and tctp protein occurs because of an association between tctp protein and heme, however, there is a lack of evidences of tctp or any other malarial protein in association with artemisinin is directly causing the parasite elimination though a variety of ultrastructural studies have been carried out on infected red cells treated with artemisinin derivatives and from those studies it was described that membrane-containing structures, such as the plasma membrane, endoplasmic reticulum, nuclear envelope, food vacuolar membrane and mitochondria appear to be most sensitive to the action of artemisinins [Maeno Y et al.1993] thought from these observations a variety of mechanisms of action might be suggested. Another very interesting observation is that when radiolabelled artemisinin derivatives are fed to malaria infected cells, the drug has been found to accumulate in hemozoin and in the membranes of the food vacuole and mitochondria [Maeno Y et al.1993]. These observations are consistent with the role of heme in the mechanism of action. Thought there are as described before many theories to explain artemisinin mode of action one cannot even say what is the cellular target for artemisinin. Tctp protein for the reasons stated before was one of the suggested targets and more recently Krishna and co-workers have indicated that artemisinins might inhibit the sarco–endoplasmic reticulum Ca 2+ -ATPase (SERCA) of P. falciparum (Pfatp6) and, therefore, also SERCAs of other Plasmodium species because thapsigargin a chemical component that is a potent and selective inhibitor of SERCAs and share some chemical similarities with artemisinins [Eckstein-Ludwig U et al. 2003]. Artemisinins, but not other antimalarial drugs, inhibited Pfatp6 protein activity when Pfatp6 was expressed and assayed in Xenopus oocytes [Eckstein-Ludwig U et al. 2003]. Others have carried out docking simulation studies of artemisinin derivatives to models of the thapsigargin binding site in Pfatp6 protein [Uhlemann AC et al. 2005]. Several potential 66 hydrophobic interactions between side chains of artemisinin derivatives and amino acids of Pfatp6 protein have been identified including Leu263 that seems to modulate artemisinin susceptibility when examined using mutagenesis experiments of malarial SERCAs [Jung M et al. 2005, Uhlemann AC et al. 2005]. Taken together, there are independent lines of evidence that have been obtained from a range of experimental techniques to suggest that Pfatp6 protein might be the primary target of artemisinins. However, it is suggested that genetic studies are required to support this hypothesis. More recently, Li and co-workers have suggested that the electron transport chain of P. falciparum might be a target for artemisinins [Li W et al. 2005]. In support of this idea, when yeast was grown in non-fermentable media (making it dependent on mitochondrial respiration), sensitivity to artemisinin increases, because over expression of some mitochondrial-transport proteins seems to increase sensitivity to artemisinins, it has been suggested that the electron transport chain stimulates the activity of artemisinins, and that these activated artemisinins impede mitochondrial function by depolarizing mitochondrial membrane potential [Li W et al. 2005]. The mechanism of this inhibition is unclear but it might be related to the presence of an iron group in the cytochrome center that induces the formation of radicals [Haynes RK et al. 2006]. 67 1.7 Drug Resistance The emergence and spread of parasite resistance to anti-malarial drugs has presented one of the largest obstacles hindering the effective treatment and control of malaria. The WHO’s official definition of malaria parasite resistance dates from 1973; “the ability of a parasite strain to survive and/or multiply despite the administration and absorption of a drug in doses equal to or higher than those usually recommended but within the limits of tolerance of the subject” [World Health Organisation. Chemotherapy of Malaria and Resistance to Antimalarials: Report of a WHO Scientific Group]. The WHO 1973 definition of the level of parasite drug resistance remains in use; A - Sensitive (S): The asexual parasite count reduces to <25% of the pre-treatment level in 48 hours after starting the treatment, and complete clearance after 7 days, without subsequent recrudescence - Complete Recovery. B - RI Delayed Recrudescence: The asexual parasitaemia reduces to < 25% of pretreatment level in 48 hours, but reappears between 2-4 weeks. C - RI Early Recrudescence: The asexual parasitaemia reduces to < 25% of pretreatment level in 48 hours, but reappears within 2 weeks. D - RII Resistance: Marked reduction in asexual parasitaemia (decrease >25% but <75%) in 48 hours, without complete clearance in 7 days. E - RIII Resistance: Minimal reduction in asexual parasitaemia, (decrease <25%) or an increase in parasitaemia after 48 hours. Though it is important to clarify that clinical treatment failure or increased in vitro IC 50 values alone are not sufficient to prove drug resistance. A parasite isolate should be classified as resistant only after analysis of treatment response parameters (parasite and fever clearance) and treatment success (determination of possible re-infection, in the case of apparent recrudescence) in correlation with the in vitro drug sensitivity. Increased IC 50 values combined with prolonged parasite clearance and treatment failure define a case of confirmed drug resistance [Noedl H 2005]. One might suggest that the first step in the development of clinical resistance may be a decrease in parasite in vitro susceptibility, associated with a key mutation in a target enzyme which ultimately results in clinical failure. Thought it is also very important to notice that the parasite response and resistance is not only dependent on the parasite genotype but also involves the general health state and immune status of the patient. A number of factors influence the likelihood of resistance occurring and the speed with which it spreads. For instance, the mechanism by which the drug works against the parasite is 68 important; simple modes of action such as simple enzyme inhibition are likely to lead to rapid evolution of resistance, as the number of genetic mutations required to alter enzyme structure is low. This is the case with pyrimethamine resistance, which evolved very quickly after the introduction of the drug, in contrast to the pattern seen with the emergence of chloroquine resistance, which took much longer to evolve, as most studies described chloroquine mode of action has being much more complex than pyrimethamine. The pharmacokinetic dynamics of drugs are also important in determining the selection pressure for drug resistance. Watkins and Mosobo [Watkins WM et al. 1993], for example, showed that the long half-life of sulfadoxine-pyrimethamine was a considerable factor in the selection pressure for resistant mutants, as the drug was present in patients at sub-therapeutic levels for long periods of time. Drugs with high efficiencies of parasite killing, rapid achievement of levels above the minimal inhibitory concentrations and short half-lives, will be the most effective at minimizing the selection pressure for resistant mutants [Winstanley PA et al. 2002]. Resistance has been recorded to every anti-malarial currently in use, with the exception of artemisinin and its derivatives. Quinine, the first drug used specifically to treat malaria was used extensively. The first reports of resistance to the drug occurred at the beginning of the 20 th century, when Couto (1908) and later Nocht and Werner (1910) reported the treatment of patients who did not respond to quinine treatment [Peters W 1987]. Despite the appearance of quinine resistance, the drug remains remarkably useful today, especially as a first line drug for treating complicated cerebral malaria. In fact, quinine resistance is surprisingly uncommon, and in the few instances it has emerged, it is often associated with parasites that are already resistant to other drugs such as chloroquine and mefloquine [Looareesuwan S et al.1990; Meshnick SR 1997; Peters W 1987; Pukrittayakamee S et al. 1994]. Chloroquine, itself based on the structure of quinine, was developed in Germany in the early 1940s, resistance to chloroquine was far more forthcoming than with quinine, and the first reports of parasites failing to respond to the drug emerged independently from South America and South East Asia in the late 1950s [Moore DV et al. 1961; Young MD et al. 1961]. The spread of resistance from these pioneer areas was relatively rapid, and chloroquine resistance is now a major problem throughout the malaria affected areas of the world. One of the proposed mechanisms for the emergence of drug resistance is through the presence of a drug at sub-therapeutic levels within a population. There can be no doubt that the emergence of chloroquine resistance in South America was facilitated by the policy of distributing chloroquinated salt to the area as part of a well-intentioned control problem. This resulted in a large proportion of the population being exposed to the drug at sub-curative 69 doses, thus considerably enhancing the chances of selection of chloroquine resistant parasites [Payne D 1988]. The proliferation of chloroquine resistance resulted in the development of mefloquine, a drug that was effective against chloroquine resistant parasites. Initial indications that mefloquine resistance was likely to emerge came, however, in 1977, when resistance was experimentally induced in a rodent malaria parasite [Peters W et al. 1977]. Efforts to reduce the possibility of the emergence of resistant parasites in the field by using mefloquine in combination with other drugs (especially pyrimethamine) met with failure, however, and reports of mefloquine resistant parasites emerged throughout the 1980s [Peters W 1998]. Introduced as a first line treatment to Thailand in 1984, substantial resistance had developed within 6 years [Price RN et al. 2004]. Concerning the genetic of drug resistance, two main genes have been implicated in chloroquine resistance; the pfmdr1 (P. falciparum multi drug resistance 1 gene) and the pfcrt (P. falciparum chloroquine resistance transporter gene). Pfmdr1 protein is a membrane protein, belonging to the sub group of ABC-type multidrug transport system. P. falciparum gene mdr1 is localized on chromosome 5 and according to the P. chabaudi synteny map the pcmdr1 (gene homologous on P. chabaudi of the pfmdr1) gene is localized on chromosome 12. Pfcrt protein is a digestive vacuole transmembrane protein, associated to chloroquine resistance. P. falciparum gene crt is localized on chromosome 7 and according to the P. chabaudi synteny map the pfcrt gene referred to as cg10 gene is localized on chromosome 6. It has been shown that some point polymorphisms in the pfmdr1 gene can be correlated with chloroquine resistance in some field isolates [Basco LK et al. 1995, Cox-Singh J et al. 1995, Duraisingh MT et al. 1997, Duraisingh MT et al. 2000]. Similarly, transfection work has suggested that the gene pfmdr1 can modulate the sensitivity to chloroquine [Reed MB et al. 2000]. However there are other studies found in the literature with parasites collected from the field where no association between point polymorphisms in the gene pfmdr1 and chloroquine resistance was found [Chaiyaroj SC et al. 1999, Cremer G et al. 1995, Povoa MM et al. 1998]. In addition, the analysis of a genetic cross between a P. falciparum chloroquineresistant clone; Dd2 and a chloroquine sensitive one, HB3, showed that mutations in the gene pfmdr1 did not genetically segregate with chloroquine resistance [Wellems TE et al. 1990]. Later detailed linkage analysis and fine chromosome mapping of the progeny clones of the P. falciparum Dd2 x HB3 genetic cross allowed the identification of another gene, the pfcrt gene, in which a particular mutation at the amino acid position 76 (K76T) a lysine to a threonine change, that appears to correlate completely with chloroquine resistance among 70 field isolates of P. falciparum [Fidock DA et al. 2000]. Babiker HA and colleagues described that the combination of mutation in both pfmdr1 and pfcrt gene loci confer higher chloroquine resistance phenotypes in field populations of P. falciparum [Babiker HA et al. 2001]. In relation to the genetic of mefloquine resistance, in vitro studies on P. falciparum have shown that genetic amplification of the pfmdr1 gene may correlate with both mefloquine and quinine resistance [Cowman AF et al. 1994, Peel SA et al. 1994]. However, field studies on the association between the pfmdr1 gene and the parasite response to mefloquine have not provided unanimous results, while some appear to have shown an association between the amplification of the pfmdr1 gene and mefloquine resistance [Price RN et al. 1997, Price RN et al. 1999, Wilson CM et al. 1993] others as Chaiyaroj SC and co-workers have not found any correlation between the two events [Chaiyaroj SC et al. 1999], one possible explanation comes from the fact that the genotype-phenotype associations depends from the parasite geographical origin. In addition and as with chloroquine it has also been demonstrated through genetic crossing and transfection experiments that point mutations in the pfmdr1 gene may modulate the sensitivity to both mefloquine and quinine in P. falciparum [Duraisingh MT et al. 2000, Reed MB et al. 2000]. In P. chabaudi Cravo PV and colleagues [Cravo PV et al. 2003] shown that amplification of the pcmdr1 gene (gene homologous of the P. falciparum mdr1 gene) is an important event in the generation of mefloquine resistance, paralleling the situation observed by Cowman AF and co-workers and Peel SA and colleagues [Cowman AF et al. 1994, Peel SA et al. 1994] though other genes are suggested to be also involved. 71 1.7.1 Resistance to artemisinin Clinical parasite resistance to artemisinin drugs has not yet been observed, although variations in sensitivity have been described [van Agtmael MA et al. 1999] for example various isolates of P. falciparum from Vietnam and Thailand have been found to vary in their sensitivity to artemisinins in vitro [Brockman A et al. 2000; Woitsch B et al. 2004; Wongsrichanalai C et al. 1997; Wongsrichanalai C. et al. 1999]. Differences in the sensitivity level of P. falciparum isolates to artemisinins can be due to genetic alterations of the parasite that confer differential sensitivity to these drugs or simply can be due to the natural genetic variation of the parasite in that particular part of the world. To this stage, several proteins, including a Ca 2+ -depending SERCA type ATPase protein that in P. falciparum is codified by the pfatp6 gene that is localized in P. falciparum chromosome 1, the P. falciparum chloroquine resistance transporter protein codified by the pfcrt gene, the P. falciparum multidrug resistance protein-1 codified by the gene pfmdr1, and the translationally controlled tumor protein (tctp) that is codified by the gene pftctp in P. falciparum (this gene being localized in P. falciparum chromosome 5), have been implicated in modulation of parasite susceptibility to artemisinin drugs. Mutations in pfcrt gene were associated with increased susceptibility of P. falciparum isolates gathered in Asia, Africa and South America [Sidhu AB et al. 2002] to artemisinin. Measurement of the pfmdr1 gene copy by real-time PCR on filed isolates that were significantly more resistant to mefloquine, quinine, artemisinin and artesunate and more sensitive to chloroquine showed that this isolates had 3 copies of pfmdr1 gene (in normal conditions pfmdr1 is a single copy gene), thus, reduced in vitro sensitivity to artemisinin and artesunate was correlated to an increased gene copy numbers of the gene pfmdr1 [Pickard AL et al. 2003, Price RN et al. 2004]. Price RN and colleagues also associated polymorphisms in the gene pfmdr1 with an increased artemisinin susceptibility in isolates with a single copy of the gene pfmdr1, in this study the N86Y mutation was associated with lower IC 50 s to mefloquine than in those isolates with a wild-type pfmdr1 gene. By contrast, the presence of either the S1034C mutation or the N1042D mutation in isolates with single copies of the gene pfmdr1 was associated with a higher artesunate IC 50 s than was the wild-type, at both these loci. On the other hand other point polymorphisms of the gene pfmdr1 have been associated with increased sensitivity to artemisinin [Duraisingh MT et al. 2000]. Likewise, the triple mutation in the gene pfmdr1, S1034C/N1042D/D1246Y, highly prevalent in South America, was found to enhance parasite susceptibility to mefloquine, halofantrine and artemisinin [Sidhu AB et al. 2005]. From all these studies the gene pfmdr1 appears to be one important 72 modulator of the parasite susceptibility to artemisinin drugs. Higher copy numbers of this gene predict treatment failure even for chemotherapy with the highly effective combination of mefloquine and 3 days artesunate [Price RN et al. 2004]. It seems that there is a correlation between the resistance to artemisinins and to other antimalarials. Since the gene pfmdr1 has a general importance in antimalarial drug resistance, induction of resistance to one drug may be followed by resistance to other drugs that are not active by the same mechanism [Anderson TJ et al. 2005; Duraisingh MT et al. 2000; Ferrer-Rodriguez I et al 2004; Ngo T et al. 2003; Pickard AL et al. 2003; Price RN et al. 2004; Reed MB et al. 2000; Sidhu AB et al. 2005]. Resistance to artemisinin in P. yoelii had been previously been selected by drug pressure and was correlated to a high protein expression level of parasite tctp protein (Translationally Controlled Tumor Protein Homolog) [Walker DJ et al. 2000], which has been shown to bind artemisinin [Bhisutthibhan J et al. 1998], however, the resistant parasites readily lost resistance once drug-selection pressure was withdrawn [Peters W et al. 1999], so no clear association was actually made between artemisinin resistance and the protein tctp, as being a transient phenotype, the genetic involved on this resistance phenotype is not possible, that is why one of the mains objectives of this project was to select artemisinin resistance of stable phenotype. So in conclusion we can say that we think that the genetic mechanism responsible for artemisinin and its derivatives resistance need clarification. To start the process of clarification it is better to clarify that although studies made directly on P. falciparum may provide more incisive information, this presents several limitations starting by the fact that in the artemisinin case there is no artemisinin resistance yet reported, so the work in animal models can circumvented all the human malaria parasite limitations. There are several rodent malaria models available; those will be presented in the following chapter. 73 1.8 Rodent malaria parasites The host specificity of human malaria parasites represents a major constraint on the study of malaria as, unlike the other major tropical diseases such as trypanosomiasis and leishmaniasis, the actual causative organisms cannot be maintained in convenient small laboratory animals. The need for suitable laboratory models has resulted in the use of avian and simian parasites and until 1948 these models were the only ones available. In 1948 the situation changed with the discovery and isolation of a malaria parasite that was capable of infecting laboratory rats and mice [Vincke IH et al. 1948]. This parasite, Plasmodium berghei, soon became the most intensively studied malaria parasite. All experiments described within this PhD thesis were carried out using the rodent malaria parasite, Plasmodium chabaudi. P. chabaudi belongs to a group of four Plasmodium species that infect murine rodents from Central Africa the other species being Plasmodium vinckei, Plasmodium yoelii and Plasmodium berghei. A map of the locations from which the various rodent parasites were isolated is shown in Figure 8. Figure 8 The geographic origins of the rodent malaria parasites. From Carlton JM et al.2001 with kind permission of Dra. Jane Carlton. 80 1.9 Linkage Group Selection As any other living organism, in Plasmodium the identification and fully understanding of the genetic mechanisms involved in important phenotypes such as drug resistance is of great importance. In the particular case of drug resistance the knowledge of the genes involved in this phenotype allows epidemiological studies and other long term studies for example as the ones monitoring the resistance of a particular antimalarial in the field. In the case of malaria parasites there are nowadays mainly two genetic methods that can be used to try to locate genes controlling any genetic trait such as drug resistance for example; those are linkage analysis and linkage group selection (LGS). When using classical linkage analysis to identify the genetic loci involved in drug resistance, the resistant mutants need to be crossed with genetically distinguishable parasites (like AS and AJ strains for example), this genetically distinguishable parasites must differ in a number of genetic markers. These markers can be restriction length polymorphism (RFLPs) or microsatelites for example, and will distinguish the sensitive from the resistant parasites. When using classical linkage analysis, the progeny of the genetic cross obtained needs to be cloned and the resulting cloned progeny will be analysed for linkage of the phenotype, lets consider for the moment drug resistance, with the inheritance of any parental markers either resistance or sensitive. Classical linkage analyses bring to notice groups of markers whose inheritance is linked to the inherence of drug resistance. In the eventual case of having a number of markers considered sufficient a linkage map can be constructed [revised by Carter R et al 2007]. Classical linkage analysis has been used before with success to identify regions of the parasite genome that are important for drug resistance as for example the case of Wellems TE and co-workers, Carlton JM and co-workers and Hunt P and collaborators [Carlton JM et al. 1998, Hunt P et al. 2004b, Wellems TE et al. 1991] but only once has it lead to the actually identification of a gene responsible for a particular phenotype, in this case, it was the identification of pfcrt has the gene involved in chloroquine resistance [Fidock DA et al. 2000, Su X et al. 1997]. Due to the fact that for using classical linkage analysis a large number of genetic markers is indeed necessary, it is also necessary to know the position of these markers in the parasite genome and then involves the cloning of a large number of clones from the recombinant progeny and also the genetic characterization of each one, this technique is very laborious and time consuming. The other technique that enables the discovery of genes controlling biological properties in malaria parasites as stated before is linkage group selection (LGS). LGS is a novel approach developed for malaria parasites in order to identify genes responsible for selectable 81 phenotypes, as for example drug resistance. LGS was previous validated for finding genes involved in drug resistance by Culleton R and co-workers [Culleton R et al. 2005]. LGS can be applied to the genetic analysis of any malaria parasite as long as there are experimental means of infecting mosquitoes with gametocytes from parasites and also means of passaging the genetic progeny of a cross through the liver stage of development and into the blood. LGS has in common to the classical linkage analysis the fact that it is an approach that uses a genetic cross between two unrelated parasites from the same species, one of which is sensitive and the other one is resistant to a particular characteristic that is going to be applied as selective pressure [Culleton R et al. 2005], but differs from the traditional approach by avoiding cloning very large numbers of cloned lines from the progeny of a genetic cross [Carter R et al. 2007]. In LGS the uncloned progeny of a genetic cross, between a sensitive and a resistant parasite to a particular characteristic and that are genetically distinguishable by a large number of genetic markers such as AFLPs markers for example, is placed under a selection pressure representing the biological property of use, in the case of our project; artemisinin and artesunate resistance. The DNA obtained from the surviving progeny is then screened from the presence of a large number of molecular markers distributed throughout the parasite genome. Prior to the developed of the LGS technique an atlas of a large number of molecular genetic markers distinguishing for example the two different strains AS and AJ was developed [Martinelli A et al 2005]. The genetic markers from the sensitive progenitor that are linked to the gene of interest (for example in our case the gene conferring artemisinin resistance) will be under-represented or even eliminated from the progeny of the genetic cross after the drug selection. Finding the genetic position of the markers under selection would allow us an area in the genome where the gene of interest might be located. It is important to noticed that the intensity of reduction of any particular marker is directly proportional to the distance of the gene of interest thus allowing the construction of a selection valley around the genetic area of interest, where the target gene in supposedly located at the valley base [Carter R et al. 2007]. See Figure 11 for the resume of the LGS protocol. 82 Figure 11 – Schematic representation of the Linkage Group Selection protocol. P Pa ar re en nt ta al l R Re es si is st ta an nt t P Pa ar re en nt t A AS S- -A AR RT T a an nd d A AS S- -A AT TN N S Se en ns si it ti iv ve e P Pa ar re en nt t B Bl lo oo od d f fo or rm ms s m mi ix xe ed d P Pr ro og ge en ny y R R e e c co om m b b i in n a a n n t t c cl l o o n n e e s s P Pa ar re en nt ta al l P Pa ar re en nt ta al l G Ga am me et to oc cy yt te es s S Se el lf fi in ng g, , C Cr ro os ss s f fe er rt ti il li is sa at ti io on n R R e e c c o o m m b b i i n n a a t t i i o o n n R Re es si is st ta an nt t P Pa ar re en nt t S Se el le ec ct ti io on n P Pr re es ss su ur re e S Se el le ec ct te ed d R Re ec co om mb bi in na an nt t P Pr ro og ge en ny y w wi it th h t th he e a al ll le el le e o of f i in nt te er re es st t 83 1.9.1. Amplified Fragment Length Polymorphism LGS requires that the two parental clones are distinguished by a large enough number of genetic markers to ensure that some will be linked to the genes of interest. LGS was optimized using amplified fragment length polymorphism (AFLP) a PCR-based method for amplifying DNA fragments from genetically distinct cloned lines of parasites, for example AS and AJ. AFLP is a technique in which large numbers of markers are generated across a genome [Masiga DK et al. 2000]. Having a high density of these markers in the genome means that there is a high probability that some of them will be linked to the gene of interest. It has been previously shown that AFLP meets the requirements of LGS both as regards numbers of markers generated in different strains of P. chabaudi [Grech K et al. 2000] and their quantisation in a mixture of the strains [Martinelli A et al. 2004]. The AFLP technique allows the visualization of restriction fragments of DNA. This enables the detection of the variation in DNA between two strains without prior knowledge of the nucleotide sequence. Depending upon the polymorphism between the two strains, large numbers of genetic markers can be produced in a relatively short time [Vos P et al. 1995]. AFLP involves cutting genomic DNA into a large number of DNA fragments with two different restriction enzymes (an enzyme called a frequent cutter and another called a rare cutter), thus generating optimal size fragments for visualization on polyacrylamide gels. Using radiation to label one of the primers allows visualization of the products. Genetic differences between strains occurring at a cutting site result in fragments of different sizes which can then be identified on a gel as being present in one strain and absent in the other (See Figure 12 for an example of an AFLP gel). Polymorphic bands can be produced as a result of polymorphisms between strains at the enzyme recognition sites, polymorphisms between strains at the selective bases used in each PCR, or as a result of insertion/deletion polymorphisms within fragments. 84 1.10 Aims of the project This project involves identifying and characterizing genes involved in resistance to artemisinin using the rodent malaria specie P. chabaudi, starting by selecting stable artemisinin drug resistance. To select artemisinin and artesunate mutants using the rodent model, P. chabaudi, though prolonged exposure of drug-sensitive lines to low and increasing levels of the drug in mice. Attempts of selecting artemisinin and artesunate mutants in P. chabaudi would be carried out though prolonged exposure of drug-sensitive lines to low and increasing levels of the drug in mice; so that the surviving parasites of one lower dose of artesunate and artemisinin would receive an increasing dose of the same drug. These results will be presented on chapter III . After selecting mutant clones resistant to artemisinin and artesunate the objective would be, as a first approach, to study the involvement of P. chabaudi gene orthologues pfmdr1, pfcrt, pftctp and pfatp6, in the selected mutant clones, previously described as being putative genetic modulators for artemisinin. These results will be presented on chapter IV . The last objective was to perform genetic crosses in mosquitoes with the previously selected and cloned artemisinin and artesunate mutants in P. chabaudi (AS-ART and AS-ATN respectively) and the genetic distinctive sensitive parasite line of P. chabaudi AJ, and to perform linkage group selection (LGS) on the genetic crosses using artemisinin and artesunate; these results will be presented on chapter V . 85 CHAPTER II MATERIALS AND METHODS 86 87 2.1. Mice Inbred female CD1 (Mus musculus) from Harlan-Tekld Iberica were used for drug tests and selection experiments. Inbred female CBA/CA and C57B1/6J mice (Mus musculus) from the University of Edinburgh were used in the rest of the experiments described in this work. All mice used were 4-6 weeks old at the start of the experiments. They were housed in polypropylene cages with sawdust as bedding, and were provided with 41B rat and mouse maintenance diet (Harlan-Tekld) ad libidum. Drinking water was supplemented with 0.05% paraminobenzoic acid (PABA), an essential element for the parasite growth, and given ad libidum. The cages were kept in a room where temperature was maintained at a constant 25º C ± 3 º C, on a 12-hour light/dark cycle. 2.2. Mosquitoes Anopheles stepehensi mosquitoes were used for all experiments. Insectaries were maintained on a 12-hour light/dark cycle at 25-27º C temperature and 75-86 % humidity. Larvae were fed on Liquifry™ until 2 nd instar and thereafter on ground Tetramin™ fish food, Adult mosquitoes were kept on 10% glucose and 2% PABA-supplemented water solution. Stock adults received weekly rat-blood feeds, which are essential for the production of eggs. 2.3. Parasites All parasites used in the experiments described here were Plasmodium chabaudi chabaudi (referred to as P. chabaudi hereafter). See Table 5 and also Table 6. Table 6 - Parasite clones and lines used in the present work. CLONE DRUG RESPONSE AJ Drug sensitive, genetically different and distinguishable from AS line AS-PYR Derived from AS-SENS; pyrimethamine-resistant AS-15CQ Derived from AS-3CQ; resistant to 6 daily doses of chloroquine at 5mg/kg mouse body weight AS-30CQ Derived from AS-15CQ; resistant to 6 daily doses of chloroquine at 30 mg/kg mouse body weight 88 2.4 Preparation of 10 7 standard parasite inocula 10 7 iRBC was established as the standard parasite number to be infected into individual mice at the time of the drug tests, during selection experiments and during preparation of inocula for the production of mixed infections. The preparation of the inocula for drug tests and for the selection experiments was as follows: the parasitaemia and red blood cell density (RBC/ml) were calculated in the donor mouse, after which the required amount of blood was collected from the mouse tail by calibrated capillary pipette and diluted to a final concentration of 10 7 iRBCs/0.1 ml in a solution of heparinised 1:1 calf serum/mammalian Ringer’s solution (Appendix 1). Preparations were kept on ice at all times and an aliquot of 0.1 ml was inoculated into mice. For the preparation of inocula for the production of mixed infections containing equal proportions of two parasite clones, infections of both clones into a single mouse were induced as described above. 2.5 Cloning Dilutions for cloning were prepared as in 2.4 except that each mouse was infected either with a mean of 0.5 or 1 parasite. Groups of 50 mice were inoculated in these experiments. If approximately 30% of mice became infected, it could be predicted that 75% of the infections had resulted from a single parasite; a lower percentage of infected mice would indicate higher proportion of pure clones. Cloning was done once drug resistance stability was verified. 2.6 Preparation and administration of artemisinin and artesunate Artemisinin powder was obtained as a gift from African Artemisia. Artesunate powder was obtained as a gift from Daphra Pharma. In the initial tests artemisinin was dissolved in dimethyl sulphoxide (DMSO) and corn oil, and artesunate was dissolved in DMSO, Na 2 CO 3 or corn oil, but for further analysis DMSO was always used. Both artemisinin and artesunate were freshly diluted daily in dimethyl sulphoxide (DMSO) and were kept at room temperature protected from the light. Both drugs were administered to mice by gavage using a lubricated catheter adapted to a 1 ml syringe. Drug doses were expressed as milligrams of drug per kilogram (Kg) of mouse body weight per day. Both drugs 89 were diluted to a concentration such that the amount of drug corresponding to the desired dose was present in 0.1 ml when given to a mouse weighing 20 grams. At the time of drugging, mice were individually weighed so that the amount of drug given could be adjusted. 2.7 Drug tests Initially, artemisinin and artesunate drug test trials were carried out on four P. chabaudi clones (AS-PYR, AS-15CQ and AS-30CQ), to establish the appropriate drug regimen and a standard test to distinguish between resistant and sensitive parasites as well as parasites with intermediate levels of resistance. Mice received 10 7 iRBC each by intra-peritoneal (i. p.) injection. Groups of five mice were prepared; one group was untreated and used as a control for the infection, while the remaining group was drugged three hours after injection, to allow parasites to reach the blood stream. Depending on the particular drug regimen the drug dose was repeated every 24 hours for the desired number of days. Blood smears from control and artemisinin or artesunate treated mice in the single dose, in the three-day suppressive test and in the five-day suppressive test, were taken on day five post infection and every day thereafter, until the infection peaked or it was clear that no parasites were going to appear. The parasitaemias and number of days for recrudescence of the different parasite lines were compared. 2.8 Artemisinin and artesunate selection experiments: general procedure Two groups of five (4-6 week old) CD1 mice were inoculated with 10 7 parasites of the clone to be used for selection for increased artemisinin or artesunate resistance. Three hours after inoculation one of the groups were treated orally with the required doses of artemisinin or artesunate for a total of five days. The remaining group was left untreated and served as a control. Both controls and parasites that survived drug treatment were passaged weekly from the mouse exhibiting the highest parasitaemia into uninfected mice and the treatment repeated. The drug doses were increased according to the parasite response to treatment in the previous passage. To address the possibility that potential increases in drug tolerance could be to due to increased virulence attributed to multiple sub-inoculations, an untreated and unselected parasite line was maintained in parallel and passaged in untreated mice the same number of times as the drug selected line. Following these passages in the 96 For the AS-ATN x AJ cross, parasites were pooled and sub-inoculated from the “nonpassaged” group into two further groups of mice, one of which was treated with artesunate (“ATN treated” group), and the other left untreated (“untreated” group). These two groups provided the material for the comparison of markers between drug-treated and untreated parasite populations. Artemisinin was administered orally at a dose of 25 mg/kg of mouse body weight daily at 24-h intervals for 5 days, starting 3h after parasite challenge. Artesunate was administered orally at a dose of 5 mg/kg of mouse body weight daily at 24-h intervals for 5 days, starting 3h after parasite challenge. Both the artemisinin-treated and artesunate-treated and the untreated blood-stage cross progeny were allowed to grow to peak parasitaemia (30%–40% for untreated, and 20%–30% for treated), at which point the blood was harvested. Two samples of parasite DNA were prepared for AFLP and other molecular analyses by pooling separately the blood from the treated and untreated mice. 2.16.2 Amplified Fragment Length Polymorphism (AFLP) analysis 2.16.2.1 Preparation of parasite DNA from experimental groups When infections reached peak parasitaemia, blood was extracted from all mice in each group by severance of the brachial artery, pooled, and prepared for DNA extraction. Blood was filtered in order to remove any mouse lymphocytes or other nucleated cells, by passing it twice through a 5 ml column of powdered cellulose (Sigma) washed with citrate saline. Blood was then filtered through Plasmodipur™ filters (Euro-Diagnostica) twice. The filtrate was centrifuged for 5 mins at 3000 rpm and the supernatant removed, leaving a pellet of packed cells. 0.15% saponin in Phosphate Buffered Saline (PBS) was added to promote cell lysis cells. After lysis of erythrocytes occurred (associated with a change in colour of the solution from bright red to burgundy colour), PBS was added in excess to prevent parasite lysis. This solution was then centrifuged again at 4000 rpm for 5 minutes and washed twice in PBS. Supernatant was discarded and pellets stored at -70°C. Three thick blood smears were taken, one prior to filtration, one after cellulose filtration and one after Plasmodipur™ filtration in order to determine the efficiency of host cell removal at each stage. The frozen pellet was re-suspended in 0.4 ml buffer A (Appendix 1), and 10 µl of 10% SDS and 50 µg Proteinase K (Sigma) were added. The pellet was left at 37°C overnight after which an equal volume of 1:1 phenol/chloroform mixture was added, mixed for 3 min, and centrifuged at 5,000 g for 1-2 minutes. The upper aqueous layer was transferred to a fresh 97 tube. The step was repeated 2-3 times. Then an equal volume of chloroform was added, and the tube was centrifuged as before for 1-2 minutes. The upper aqueous layer was removed to a fresh tube. This procedure was repeated once before an equal volume of ether was added, the solution centrifuged as before for 1-2 minutes and the upper layer removed. The remaining ether was left to dry in the air. Three volumes of absolute ethanol (0º C) and 1/10 th volume of 3M sodium acetate (pH 5.2) were added to the dried pellet, and the tube was mixed and placed on ice for 15-45 minutes to precipitate the DNA. The tube was then centrifuged at 10,000 g in a Speed Vac (Savant) for 10 minutes and the ethanol mixture removed. The tube was again centrifuged at 10,000 g for 5-10 minutes to remove final traces of ethanol and the pellet was resuspended in 100-200 µl of TE buffer (pH 8.0) (see Appendix 1) and left at 37°C for 10 minutes before storage at -20º C. 2.16.3 AFLP analysis 0.5µg of parasite genomic DNA was cut using two enzymes. Firstly, 10 U of EcoRI (MBI Fermentas, recognition sequence: G↓AATTC) were added and DNA incubated at 37°C for 1 h, then 5 U of TruI/MseI (MBI Fermentas, recognition sequence: T↓TAA) were added and DNA incubated for further 3 h at 65°C. The digestion stage was performed in a 40 µl solution containing 2X Y+/ Tango buffer (Promega). The fragments were then ligated with adapters matching the cut ends produced by the enzymes. All primers used were provided by MWG-Biotech UK Ltd. Adapters disrupted the cutting site recognised by the enzymes in order to prevent cutting of the adapters from the DNA fragment. These adapters also provided a recognition site for primers. -MseI adapters: MeI.a1 and MeI.a2 -MeI.a1: 5’-GACGATGAGTCCTGAG-3’ -MeI.a2: 3’-TACTCAGGACTCAT-5’ -EcoRI adapters: EoI.a1 and EoI.a2 -EoI.a1: 5’-CTCGTAGACTGCGTACC-3’ -EoI.a2: 3’-CATCTGACGCATGGTTAA-5’ 98 Bold letters indicate base substitution to disrupt enzyme cutting site. Adapters were prepared by adding equimolar amounts of both strands and then performing the following procedure: - Heat adapters at 94°C for 5 mins, then cool to 21°C for 5 mins - Heat adapters at 72°C for 60 s., then cool to 21°C for 5 mins - Heat adapters at 65°C for 60 s., then cool to 21°C for 5 mins. Adapters were then diluted to 50 pmol/µl. 10 µl of ligation mixture were then added to 40 µl of digested DNA and this incubated at 37°C for 3h, then overnight at 15°C. Ligated material was diluted 1:10 in TE buffer (pH 8.0) (Appendix 1) and stored at -20°C. A preliminary PCR amplification was performed involving the use of “non-selective” primers (meaning primers with no extra nucleotides added at their 3’ ends extending beyond the adapters sequence) matching the adapters: - Non-selective EcoRI primer: 5’-GACTGCGTACCAATTC-3’ - Non-selective MseI primer: 5’-GATGAGTCCTGAGTAA-3’ A 20µl PCR solution containing 0.32 µM of each of the non-selective primers, 1 µl template DNA, 0.4 U Taq polymerase (Promega), 1X Mg-free PCR Buffer (Promega), 1.5 mM MgCl2, and 0.2mM of all 4 dNTPs was set up. The following cycles were performed: - 94°C for 60 s., then - 94°C for 30 s. - 56°C for 60 s. - 65°C for 60 s. repeat the three steps for 20 cycles The resulting PCR material was diluted 50-fold in TE buffer (pH 8.0) and stored at - 20°C. Selective amplification was performed using radiolabelled primer (EcoRI-primer) and selective primers (meaning primers with a selective extension at the 3’-end, in order to reduce the number of fragments amplified). For radiolabeling, 2.5x Kinase buffer (Promega), 20U T4 polynucleotide Kinase (Promega), 100 µCi [γ33 P] ATP / [γ32 P] ATP (ICN) and 500 ng of oligonucleotide primer were incubated in a 20 µl solution at 37°C for 60 min. The reaction 99 was stopped by adding 1 µl of 0.1 M EDTA pH 8.0 and heating at 70°C for 10 min. The mixture was then made up to a volume of 50 µl by adding sterile, distilled water. Primer purification was performed using TE Micro Select-D, G-25 microcentrifuge spin columns produced by Eppendorf-5 Prime Inc. Hot PCR with the radiolabelled primer was performed as follows: 0.32 µM of the selective MseI primer, 0.05 µM of the labelled selective EcoRI primer, 1 µl template DNA, 0.4 U Taq polymerase (Promega), 1X Mg-free PCR Buffer (Promega), 2.5 mM MgCl2, and 0.2mM of all 4 dNTP’s were added to a 20µl total volume PCR solution. PCR conditions for this stage were: - 94°C for 60 s., then - 94°C for 30 s. - 65°C for 60 s., annealing temperature is reduced at each cycle by 0.7°C for the next 12 cycles, then remained at 56°C for the remaining 23 cycles - 65°C for 60 s. repeat the preceding three steps for 35 cycles The PCR products were mixed with 20 µl of loading dye specific for poly-acrylamide sequencing gels (Anachem), then heated at 99°C for 3 min and immediately cooled on ice. 5µl of each sample was loaded onto a 5% denaturing polyacrylamide gel (5% acrylamide, 0.25% methylene bisacryl, 7.5 M urea in 50mM Tris/50mM Boric acid/1mM EDTA). 500 µl of a 10% Ammonium Persulfate solution (APS) and 100µl of TEMED (Sigma) were added to 100 ml of gel solution and the gel cast using a SequiGen 38x50 cm gel apparatus (BioRad). Electrophoresis was performed at 110 W for 2 h in 1x TBE buffer. Gels were then dried in a vacuum gel drier (model 583, Bio Rad) and exposed overnight in phosphorimager screens (Fuji) at -70°C. Results were visualised on an autoradiography film (Kodak XAR-5). They were developed in an automatic autoradiographer developer (Exograph). As an example of an AFLP gel and the band nomenclature see Figure 12. Thus, and just as an example, the marker AJAG02CA denotes the second largest (that is where the 02 cames from) AJ-specific band (thus beginning with AJ) obtained using EcoRI primers with AG (the first pair of nucleotide letters, like the forward primer) as additional “selective” 3’-nucleotides, and CA as “selective” nucleotides on the MseI primers (the second pair of nucleotide letters, like the reverse primer). 100 Figure 12 - An example of an AFLP gel. In this particular gel we can see the results of bands generated with Mse1 (CA selective bases, so called forward primer) + EcoR1 (AG selective bases, so called reverse primer). One of the markers of the sensitive parent (AJAG02CA) was absent in the treated population and is then presumably closely linked to the locus conferring resistance. Each one of the lanes is labelled as followed; AS: denotes DNA extracted from single AS-SENS clone infection (in our project the AS alleles are the resistant ones). AJ: denotes DNA extracted from single AJ clone infection (in our project the AJ alleles are the sensitive ones). 1: denotes DNA extracted from the unpassaged pooled group. 2: denotes DNA extracted from the treated pooled group (in the case of our project the AS-ART x AJ cross was treated with ART and the AS-ATN x AJ cross was treated with ATN). 3: denotes DNA extracted from the untreated pooled group. Each lane represents a DNA sample obtained from the pooled DNA of all mice on each experimental group. Polymorphic markers are bands that are different for AS or AJ. AS specific markers (thus resistant) are marked with a broken arrow (), and AJ specific markers (thus sensitive) with an open arrow (), bands marks with an arrow () represent a non-polymorphic marker (there are no differences between AS and AJ within this marker). Adapted from Culleton R et al. 2005 with kind permission of Dr Richard Culleton. 101 2.16.4 Measurement and comparison of the intensity of AFLP markers This technique makes uses of a large number of molecular genetic markers distinguishing two strains of malaria parasites, the AFLP markers. These markers can be visualised as bands on a polyacrilamide gel. Most of the bands that will appear on a gel will be present in both the parasites and are called non-polymorphic bands and some are unique of one of the clones in our case either AS - resistant or AJ - sensitive. Once the selection pressure is applied in the genetic progeny of a cross, in the case of this project, artemisinin or artesunate pressure, the parasites that carry the allele that is sensitive (in this case the AJ allele) will be removed, and from an experimental point of view this disappearance is visible by the decrease in intensity or complete disappearance of an AFLP band specific for the sensitive allele (AJ band). Thus the progeny after selection is then screened for markers. Specially those from the sensitive parent that would either be significantly reduced or absent because these markers should be linked to the locus or loci under selection, in our case under artemisinin or artesunate selection. Polymorphic markers (between AS-ART and AJ and between AS-ATN and AJ, here simplified to AS and AJ), meaning those that are different between AS and AJ, were named and placed in the AFLP map, to denote specificity of the polymorphic band meaning the size of the band (relative to other polymorphic bands in the same gel lane) and the selective bases used, this would make it very straightforward the identification of a particular band to the AFLP map already developed by Martinelli A and collegues [Martinelli A et al. 2005]. As stated before we were looking for the disappearing of AJ sensitive markers on the treated group (meaning artemisinin or artesunate treated group) in comparison of the untreated group, each one of these markers decreased in intensity or disappeared is predicted to lay in what we call a selection valley that can then be associated to a particular gene underling the resistant phenotype or at least responsible for given so type of selective advantage over the sensitive parasites. An example of an AFLP gel is represented in Materials and Methods Figure 12. As described in the materials and methods chapter, each marker band intensity was measured with PhosphorImager and IMAGEQUANT software (Molecular Dynamics). For each marker of interest, either AS or AJ, an intensity index (II) was calculated by taking the intensity of a polymorphic marker (either AS or AJ) and compare it to a no polymorphic marker, making sure that this is done using the same PCR material applied in the same polyacrilamide gel. Each II is then converted to a relative intensity index (RII) either unselected RII u or selected RII s. The RII u is calculated by making a ration between the II obtained for the unselected 102 material and the II obtained for the parental (either AS if we are looking at an AS specific markers or AJ is we are looking for an AJ marker). The RII s is calculated by making the ratio between the II obtained for the selected material and the II obtained for the parental (either AS or AJ) [Martinelli A et al. 2004]. To compare between different markers a comparative Intensity (CI) of the polymorphic markers was calculated. This comparative intensities are defined as the RII of an AFLP marker in the cross progeny selected in “treated” mice (RII t ), divided by the RII of the marker of the cross progeny grown in a parallel “untreated” group of mice (RII ut ), and expressed as a percentage: CI = (RII t /RII ut ) x 100. Keeping in mind that the objective is to identify the genomic loci under drug selection either by artemisinin or artesunate that would correspond to a selection valley associated to the resistant phenotype, AFLP markers with low CIs were identified and their position according to the previously define genetic linkage map [Martinelli A et al. 2005] were noted. 2.16.5 Assignment of AFLP markers to locations in a P. chabaudi genetic linkage map. Markers had previously been ordered on a genetic linkage map of P. chabaudi [Martinelli A et al. 2004, Martinelli A et al.2005] obtained from previously generated crosses between ASderived clones and AJ strains of P. chabaudi [Carlton JM et al. 1998; Rosario VE 1976; Walliker D et al. 1975]. A total of 674 AFLP markers were typed for each uncloned progeny analysed in this work, and were subsequently assigned to linkage groups using the previously generated map [Martinelli A et al. 2005]. A total of 674 AFLP markers were typed for each of 28 cross-progeny clones, and were subsequently assigned to linkage groups using the Map Manager QTX software [Manly KF et al. 2001]. A total of 44 RFLP markers characterized in a previous study [Carlton JM et al. 1998] were used as genetic anchors to allow the assignment of the various linkage groups to chromosomes. In total, 11 chromosomes could be identified, while 12 linkage groups of as yet unknown assignment remain, which include the three remaining chromosomes (chromosomes 2, 4, and 14). 103 2.16.6 Sequencing of AFLP markers, and their location on the Plasmodium falciparum genome AFLP bands that appeared to be under selection were excised from acrylamide gels using a sterile scalpel. The gel fragments were then soaked in an Eppendorf tube in 50 µl of autoclaved, distilled water (sdH 2 O) overnight. Gel slices were centrifuged at 12000g prior to removal of the liquid phase containing the DNA. DNA was precipitated using 1/10 volume of 3M sodium acetate (pH 5.2) and 3 volumes of ice-cold absolute ethanol. The solution was placed at -20°C for at least 1 h. Thereafter the tubes were spun for 30 minutes at high speed. The liquid phase was removed and the DNA pellet washed twice in 70% ethanol, before air drying. The pellet was then dissolved in 50 µl TE buffer (pH 8.0). The extracted DNA fragment was amplified by PCR, using the same selective AFLP primers and PCR conditions that produced it. Sequencing reactions were carried out using the protocol described, below in section 2.16.7. The sequences of markers were then physically mapped in the P. falciparum genome (sequence data for P. falciparum were obtained from the Sanger Centre website, which can be accessed at www.sanger.ac.uk_Projects_P_falciparum), using BLAST searches. 2.16.7 AFLP band sequencing and purification of PCR products Sequencing PCR reactions were set up using the ABI PRISM Big DyeTM Terminator Cycle Sequencing Ready Reaction Kit, (PE Applied Biosystems). PCR reactions were prepared to a final volume of 10µl, containing the following reagents: 4µl template DNA (100-250ng), 4µl Terminator Ready Reaction Mix, 1µl primer (3.2pmol), 1µl sterile distilled water Sequencing PCR conditions were as follows; Denaturing 95°C for 30 seconds Annealing 50°C for 20 seconds Extension 60°C for 4 minutes Following the sequencing PCR, the products were purified by precipitation with sodium acetate and 95% ethanol, and washed in 70% ethanol. x 25 cycles 104 For sequencing of purified PCR products, internal primers covering the extreme 5’ and 3’ ends of the fragments were used. All fragments were sequenced in opposite directions. Sequencing results were analysed using the SeqED V 1.0.3 software (Applied Biosystems Inc., 1992). The programme allows the visualisation of chromatograms of the sequenced DNA. So as a summary: traditional linkage analysis of individual cross progeny clones, is labourintensive and expensive, as it involves the genotypic and phenotypic characterization of individual clones from a cloned progeny of a genetic cross. Without the generation of an extremely large number of recombinant clones it has a poor resolution, which makes the actual identification of the underlying genes extremely difficult unless strong candidates are already suspected. There is an inverse relationship between the size of the locus within which possible target genes may be located and the number of recombinant clones that must be generated [Wellems TE et al. 1991]. LGS, used in this project, has enabled the discovery of genes controlling biological properties in malaria parasites to be greatly accelerated [Culleton R et al. 2005; Martinelli A et al. 2005]. LGS differs from the traditional approach to genetic analysis of malaria parasites by not requiring individually characterisation of very large numbers of clones from the progeny of the cross [Carter R et al. 2007]. LGS characterises the uncloned progeny of a genetic cross (between a resistant and sensitive parasite to the particular parasite characteristic in study) by measuring the proportion of parental polymorphic markers at genome-wide loci (AFLP), before and after drug treatment. Markers from the sensitive parent that are linked to the gene underlying the resistance phenotype will be under-represented or eliminated after drug treatment, forming a “selection valley”. In the case of this project, the AJ markers correspond to the markers from the sensitive parent; those markers will be under represented or eliminated after artemisinin and artesunate treatment. As Carter R and colleagues so clearly summarise [Carter R et al. 2007], LGS analysis depends upon several distinct experimental components. A) The crossing of two genetically distinct lines of malaria parasite by preparing a mixed infection of gametocyte-producing blood-stage parasites in a mouse, allowing mosquitoes to feed upon the mixture, with consequent parasite development and invasion of sporozoites to salivary glands; 105 B) The subjection of the progeny of the cross to a selection pressure (in the case of this research, ART or ATN pressure). This drug selection is applied to the blood stage parasites in mice infected with the cross progeny sporozoites. C) The screening of the uncloned, selected cross progeny with quantitative markers, such as quantitative real-time PCR, quantitative AFLP and proportional sequencing. So before the LGS technique could be applied the development of quantitative markers covering as much the parasite genome as possible is necessary. Prior to this project quantitative AFLP were optimized and developed for being use with AS and AJ parasite lines from P. chabaudi [Culleton R et al. 2005; Martinelli A et al. 2005]. The location of the genetic markers in a Plasmodium genome database with the expectation that markers linked to genes controlling the target of drug selection will form a selection valley containing the locus of the selected genes. A genetic linkage map, in the case of this project already previous developed by Martinelli A and co-workers [Martinelli A et al. 2005], a genome sequence database, in the case of our model P. chabaudi sequence though not fully complete with a big coverage and a complete syntenic map developed prior to this project [Kooij TW et al. 2006] allowed us to map genes and markers genetically and physically, thus allowing identification of loci under selection. Then mutations in suspected candidate genes within the locus can be identified by comparative sequencing of genes from both the resistant mutant and its sensitive progenitor [Carter R et al. 2007]. 2.17 Experiments with genetic crosses of AS-ART and AJ or AS-ATN and AJ To analyse the efficacy and composition of each of the genetic crosses and the progression of the drug selection proportional sequencing was used. Proportional sequencing was used with DNA samples from sporozoites and mice blood samples before and after drug selection. 2.17.1 Blood collection for Proportional sequencing analyses 5 µl of blood were removed from each infected mice daily using a glass capillary. The blood was placed in an Eppendorf tube which contains 2 drops of citrate saline solution. The samples were then spun in a microcentrifuge at 10,000 rpm for 2 minutes, and the supernatent removed. The resulting pellet was frozen at -70°C. 112 2.18. Amplification and sequencing of the ubp-1 gene of P. chabaudi The amplification and sequencing of ubp-1 gene has done using the same protocol as the one described for mdr1, cg10, tctp and atp6 genes, and this protocol is already described in 2.13. Genomic DNA was used as template in 50µl PCR reactions, containing 0.2µM of each oligonucleotide primer, 1x PCR buffer (Promega), 2.5 mM MgCl 2 , 0.2mM dNTPs and 0.025U/µl of Taq DNA polymerase. These were used in PCR amplifications of AS-15CQ, AS-30CQ, AS-ATN and AS-ART. Negative controls were also prepared, which contained 1µl of sterile distilled water in place of template DNA. Positive controls were prepared using a previously amplified DNA template. The oligonucleotide primers sequence is presented in Appendix 4. All PCR reactions were carried out using a UNO-Thermoblock machine (Biometra). PCR products were run on a 2% agarose gel in TBE solution and visualized under UV. Products were purified using the QIAquick PCR Purification Kit from QIAGEN and sequenced using BigDye chain termination v3.1 (Applied Biosystems). The sequencing reactions were analysed by Macrogen®. The primers used in sequencing reactions were those used for the initial amplification of the fragments. Gene and predicted amino-acid sequences were manually compiled, and then compared between drug selected and unselected clones using an internet-based interface denoted Multiple Sequence Alignment with hierarchical clustering [Corpet F et al. 1998], using default alignment parameters (http://prodes.toulouse.inra.fr/multalin/ multalin.html). 113 RESULTS 114 115 The results of the present work are divided into three chapters: • Chapter III describes the results of experiments designed to select artemisinin and artesunate resistant parasites. These experiments also involved evaluation of certain resistance features such as stability. • Chapter IV describes the results of experiments designed to analyse the involvement of previously described putative genetic modulator for artemisinin and artesunate resistance, the P. chabaudi genes mdr1, cg10, tctp and atp6. • Chapter V is concerned with the results of experiments on genetic crosses between an artemisinin resistant (AS-ART) and a sensitive (AJ) cloned strain of P. chabaudi and an artesunate resistant (AS-ATN) and a sensitive (AJ) cloned strain of P. chabaudi. Presented here are also the results of LGS experiments conducted with the uncloned recombinant progenies of the genetic crosses. 116 117 RESULTS CHAPTER III EXPERIMENTS FOR DRUG SELECTION OF ARTEMISININ AND ARTESUNATE RESISTANCE 118 119 This chapter presents the results of experiments to select artemisinin and artesunate resistance in P. chabaudi, of stable phenotype, of cloning the parasite lines obtained and in the characterization of the resistance obtained, including its stability. 3.1. Introduction The objective of this part of the project was to select artemisinin and artesunate mutants in the rodent malaria P. chabaudi, through prolonged exposure of drug-sensitive lines to low and increasing levels of the drug, administered to mice. Table 7 depicts the maximum doses tolerated by the progenitor parasite lines as established in preliminary work. Table 7 – Maximum doses tolerated by the progenitor parasite lines used to select for resistance. P. chabaudi Maximum dose for artemisinin (mg/kg/day) Maximum dose for artesunate (mg/kg/day) AS-15CQ 1.2 2 AS-30CQ 4 1.6 120 3.2 Artemisinin drug selection P. chabaudi clone AS-30CQ was exposed to gradually increasing concentrations of artemisinin, during several consecutive passages in mice, starting with a drug dose of 4 mg/kg/day, previously determined to be subcurative for these parasites (data not shown). See Table 7. The drug selection procedure is described in more detail in the Materials and Methods Chapter (see Point 2.8) and summarized in Figure 15. Figure 15 – A schematic representation of the artemisinin selection procedure. AS-ART* is uncloned. AS - 30CQ parasite clone Sensitive to artemisinin 4 mg/kg/day 15 blood passages 60 mg/kg/day Drug dose AS - ART * parasite line Resistant to artemisinin 121 Figure 16 represents the stepwise increase in drug dose during the artemisinin selection procedure. Blood passage number and artemisinin dose 0 10 20 30 40 50 60 70 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Number of sucessive blood passage Artemisinin dose (mg/kg/day) Drug dose (mg/kg/day) Figure 16 – The increase of the artemisinin dose (mg/kg/day), tolerated by AS-30CQ clone, during 15 blood passages. Figure 16 shows the stepwise increment on the artemisinin dose during selection, starting by a 4 mg/kg/day dose of artemisinin, this dose being the maximum dose tolerated by the initial parasites, AS-30CQ. The drug resistance selection process lasted for approximately 27 weeks (approximately 7 months); meaning that during this period the P. chabaudi parasite clone AS-30CQ was kept under increasing drug pressure until a concentration of 60 mg/kg/day, at the end of which resistance seemed to have been selected. Due to time constrains the selection process was finished at this stage where stability of resistance was assayed. In the first four passages under drug pressure, the total increment on the artemisinin dose was of only 6 mg (from first passage to passage number 10) only from passage number five onwards we were able to increase 5mg of ART on each passage. 128 3.4 Cloning of resistant parasites At the end of the selection procedure, it was considered that a significant level of resistance to both to artemisinin and artesunate had been obtained. The N-fold level of resistance was calculated. The artemisinin resistant parasite line AS-ART* was 15 fold more resistant than the initial clone AS-30CQ; the artesunate resistant parasite line AS-ATN* was 6 fold more resistant to ATN than the initial clone AS-15CQ. Both parasite populations were cloned by the limiting dilution method. (See Section 2.5 for details). The cloning procedure is extensively described at Rosario V 1976 . From the cloning procedure of the parasite line AS-ART*, from the 50 mice inoculated, five mice developed patent parasitemias (10 % infection rate ) from which parasites were harvested and frozen in liquid nitrogen From the cloning procedure of the parasite line AS-ATN*, from the 50 mice inoculated, seven mice developed patent parasitemias (14% infection rate) from which parasites were harvested all frozen in liquid nitrogen. Of these, one clone from each line was chosen on the basis of their faster growth rate, for further studies, including stability assays and genetic crosses and analysis. The artemisinin and artesunate resistant clones were designated P. chabaudi AS-ART and AS-ATN respectively. See Figure 21 for details. Figure 21 – A schematic representation of the clones and parasite lines of Plasmodium chabaudi used in this project. AS - SENS AS-PYR AS-3CQ AS-15CQ AS-30CQ PYR resistant, CQ sensitive CQ resistant (low) CQ resistant (intermediate) CQ resistant (high) AS-ATN AS-ART Artesunate drug selection and cloning Artemisinin drug selection and cloning 129 AS-ART and AS-ATN were tested for their response to artemisinin and artesunate respectively, immediately after cloning. AS-ART and AS-ATN clones were inoculated into naïve CD1 mice and then re-tested for their susceptibility to artemisinin and artesunate, by comparing their response to each drug with that of non-treated and passaged the same number of times as the parasites under selection, control parasites, AS-30CQ and AS-15CQ respectively. AS-ART and AS-ATN were shown to retain the same phenotype as that of the drug-resistant population from which they had been derived, under the same drug pressure of respectively 60 mg/kg/day treatment for 5 days with artemisinin, or a 12 mg/kg/day treatment for 5 days with artesunate. These parasites were then used in subsequent studies to investigate further whether the observed drug resistance was stable. These results will be presented on Section 3.5. 130 3.5 Drug resistance stability tests Once resistance to artemisinin and artesunate were selected and cloned stability of the resistant phenotype, under absence of drug pressure was assayed. Drug-resistant parasite clones, AS-ART and AS-ATN were re-tested for their drug responses after each of three different procedures: i) Freeze-thawing cycles in liquid nitrogen, ii) 12 continuous sub inoculations in mice in the absence of drug treatment and iii) Transmission through Anopheles stephensi mosquitoes into new mice. N-fold resistance index was calculated as it has been previously described. 3.5.1. Resistance stability after liquid nitrogen preservation (deepfreezing) To check for resistance stability after liquid nitrogen preservation, resistant clones were frozen down in liquid nitrogen and then thawed, and re-tested for resistance. Briefly infected blood is drawn, when the majority of asexual parasites are at the ring stage of development, into heparinised syringes. The red blood cells are pelleted by centrifugation (2000 rpm for 5 minutes) and the majority of the plasma is removed, two volumes of deepfreeze solution (Appendix 1) are added to one volume of the packed red blood cells with constant mixing in a drop wise fashion. Such material, inoculated i. p. into a mouse, after any period of storage, induced an infection, which became patent from 7-14 days after inoculation. The parasites growing in this fashion were evaluated for their response to artemisinin or artesunate in comparison with unselected control parasites, the same way as during the drug resistance selection. AS-ART and AS-ATN parasite clones phenotypes drug resistance remained unchanged in both cases. The two cloned lines retained a 15 and 6 times Nfold resistant phenotype respectively. The parasites response to the drug was the same after freezing and thawing. Stability of resistance was therefore, confirmed. 131 3.5.2. Resistance stability after blood passages in the absence of drug pressure AS-ART and AS-ATN cloned parasites were subjected to 12 passages in untreated mice, after which they were tested for their drug responses. Briefly, 10 7 infected red blood cells were established as the standard parasite number to be infected into individual mice. Infections in the absence of drug pressure were maintained by passaging resistant parasites twelve times in CD1 mice i. p. route. Treatment was 60 mg/kg/day for 5 days for ART and as 12 mg/kg/day for 5 days for ATN. Once drug tests were preformed after the twelve passages in the absence of drug treatment, it was verified that the resistant phenotype maintained because resistant parasites recrudesced as expected while sensitive parasites did not. Resistant clones retained resistance to the corresponding drug. 3.5.3. Resistance stability after cyclical transmission through mosquitoes After confirming the stability of the resistance phenotype after cloning, freezing and thawing of these clones, in the absence of drug pressure, it was necessary to demonstrate that this genetic trait was transmissible through mosquitoes. The procedure for cyclical transmission through mosquitoes was previously described by Landau and colleagues [Landau I et al. 1966].Briefly, infected splenectomised rats, in which gametocytes were present, were exposed to mosquitoes, which had been starved for 24-48 hours. The infected rodents were exposed to mosquitoes for a varied length of time (1-2 hours) depending on the feeding performance of the mosquitoes. Seven to ten days later a small number of mosquitoes were dissected in order to count the number of oocysts, which had developed on each midgut. Fifteen and, eventually, seventeen days after the blood meal, when sporozoites were present in the salivary glands, an uninfected mouse was exposed to the mosquitoes for transmission of the malaria infection. Patent blood infections could be detected after 4-8 days in blood smears from the mice. Resistant clones successfully transmitted in this fashion were tested for their response to artemisinin or artesunate in comparison with unselected control parasites. Mice infected with AS-ART or AS-ATN were used to feed A. stephensi mosquitoes. AS-ART was successfully transmitted through mosquitoes on two separate occasions, and the resulting 132 blood forms that developed in mice were renamed AS-ARTA and AS-ARTB. The N-fold resistance index of AS-ARTB were then assessed in parallel to the, untreated but passaged control line, AS-30CQ (this line was also transmitted through mosquitoes at the same time, as a control). AS-ARTB showed an N-fold resistance index of 15-fold to artemisinin relative to AS-30CQ. Thus, artemisinin resistance remained stable after transmission of the resistant parasites through mosquitoes. AS-ARTA was not tested and the parasites where deep-frozen. For simplicity and clarity on further analysis during this project, AS-ARTB clone was renamed AS-ART. In a similar fashion, AS-ATN was also subjected to mosquito transmission and showed a 6fold increase in the N-fold resistance to artesunate relative to sensitive control; therefore we also consider artesunate resistance to be stable after mosquito transmission. See Figure 22 for a summary of the drug selection / stability tests procedure. Figure 22 – A schematic representation of the artemisinin and artesunate selection procedure. Clone AS-30CQ and parasite line AS-15CQ were passaged in the absence and presence of gradually increasing doses of drug (artemisinin and artesunate, respectively). Initial drug sensitivities (4 mg/kg/day or 2 mg/kg/day) decreased (to 60 or 12 mg/kg/day) after 15 or 14 passages, respectively. The drug responses after cloning, passage in the absence of drug (“untreated”) and transmission through mosquitoes remained unchanged. Control selection procedures in the absence of drug are also shown. Adapted from: Afonso A et al. 2006. The artemisinin and artesunate resistance phenotypes were unaltered after passaging in the absence of drug pressure, after freezing and thawing and after transmission through laboratory mosquitoes (Figure 22). Thus artemisinin and artesunate resistance obtained in our rodent model P. chabaudi is stable, indicating that resistance is likely to be genetically encoded. AS-30CQ AS-15CQ 4 mg 2 mg 60mg 12mg CLONING [ Drug ] sensitive sensitive AS -ART AS -ATN 14 passages 15 passages resistant resistant 12 passages 12 passages Untreated resistant resistant AS - ARTA Not tested AS - ARTB Resistant to 60mg AS - ATN Resistant to 12mg AS - 30CQ sensitive AS - 15CQ sensitive 133 3.6 Test for cross-resistance between artemisinin and artesunate clones In order to evaluate whether the mechanisms of resistance to artemisinin and to artesunate share similar features, the responses of AS-ART to artesunate and AS-ATN to artemisinin were tested, resistant clones were cross tested with each other’s selective drugs. Artemisinin at chosen dose was given to the artesunate resistant clone (AS-ATN) and artesunate at chosen dose was given to the artemisinin resistant clone (AS-ART). For these tests several drug doses were initially tested for both resistant clones, the maximum dose of artemisinin tolerated by the artesunate resistant clone (AS-ATN) is 32 mg/kg/day for five days and the maximum dose of artesunate tolerated by the artemisinin resistant clone (AS-ART) is 8 mg/kg/day (data not shown). AS-ART showed a five-fold increase in the MCD to artesunate relative to AS-30CQ while AS-ATN showed a greater than ten-fold increase in the MCD to artemisinin, relative to AS-15CQ. These tests therefore revealed that both clones showed cross-resistance and that in both parasites there were greater increases in artemisinin resistance (15–26 fold) than for artesunate resistance (5–6 fold). The demonstrated cross-resistance suggests some sharing of genetic or pathways features in the parasites clones described here. Thought the existence of cross-resistance suggests that resistance to artemisinin and artesunate do share similar features, at least in the parasites clones described in this project. Table 10 - N-fold resistance of P. chabaudi AS-ATN and AS-ART. The absolute and relative (N-fold) drug sensitivities of AS-ATN and AS-ART after blood passage in the absence of treatment, freeze/thaw and mosquito transmission are given. MCD – minimum curative dose. From: Afonso A et al. 2006. P. chabaudi MCD ATN (mg/kg/day) MCD ART (mg/kg/day) N-fold ATN N-fold ART AS-15CQ 2 1.2 - - AS-ATN 12 32 6 26 AS-30CQ 1.6 4 - - AS-ART 8 60 5 15 134 3.7 Discussion Drug resistance genetic basis can be study using various methodologies. A correct and complete method to study drug resistance genetics depends on firstly selecting drug resistance mutants of stable phenotype, because only the presence of a stable phenotype guarantees that resistance is genetically encoded and not a physiological adaptation to the constant presence of drug pressure. Drug resistance can be selected in vitro using for example chemical mutagenesis or in vivo by drug pressure either using a single very high dose like the method used to select PYR resistance or by progressively increasing the drug concentration. The latter situation is the one that most closely mimics the actually scenario taking place in natural parasite populations. Therefore, an appropriate method that has been used for selecting and understanding drug resistance is the utilization of genetically stable resistant mutants selected through drug pressure, originated from cloned sensitive parasite lines. Both the original drugsensitive and the selected drug-resistant parasites should be genetically identical, (or isogenic), except for any mutations involved in resistance; such mutations can then be pinpointed using different approaches. This project represents the first study where malaria parasites with genetically stable and transmissible resistance to the antimalarial drugs artemisinin and artesunate were selected and used in genetic crosses. The reasons for obtaining stable resistance where others have previously tried unsuccessfully are not clear. One reason could be the use of the rodent model P. chabaudi, which had been used before to select for pyrimethamine, chloroquine and mefloquine resistance. The existence of parasite clones with an accumulation of drug resistance genetic markers may create a favourable background for the selection of artemisinin and artesunate resistance. Two parasite lines were selected and cloned: AS-ART is 15 times more resistant to artemisinin than its progenitor parasite line, AS-30CQ and AS-ATN which is 6 times more resistant to artemisinin than its progenitor parasite line, AS-15CQ. In this work the utilization of approximately the same number of blood passages, during a similar time period under drug pressure, produced an N-fold resistance for artesunate inferior to the N-fold resistance for artemisinin and this is not explainable. Metabolization differences between the two drugs may have had an intrinsic difference in the selective pressure within the host, but drug concentration studies were not carried out. A very interesting finding during this part of the project was the cross resistance between the selected resistant clones, giving a very strong indication that resistance to ART and ATN share similar features. 135 The successful selection of these parasites and the fact that they can be transmitted through Anopheles sp. mosquitoes is a significant achievement for two main reasons: a) these observations demonstrate that malaria parasites are genetically and biologically capable of sustaining stable resistance to artemisinins which had been selected through drug pressure. Consequently, it is also conceivable that, in human malaria, artemisinin resistance may appear in the future due to extensive and/or inappropriate drug usage; once it does, it may spread in the parasite population and become established; b) the resistant P. chabaudi parasites reported here could then be used to investigate the genetic determinants of resistance to these drugs. It was possible to generate ART or ATN resistance from two different parasite clones, AS30CQ and AS-15CQ which were already resistant to chloroquine and pyrimethamine. Efforts to generate ART or ATN resistance from the very original chloroquine sensitive clone (ASPYR), were abandoned, because this clone showed, under the same drug pressure methodology, high susceptibility to the treatments (data not shown). As stated before, this may be a result of a genetic ability, if not potentiation of the parasite ability to generate mutations in response to drug treatment (called the “Accelerated Resistance to Multiple Drugs” (ARMD) phenotype) [Rathod PK et al. 1997] which might have occurred during the generation of previous drug selective methods, including chloroquine. Alternatively, it is possible that the ART resistance phenotype is only expressed in chloroquine resistant clones, almost as resistance to CQ is required for the selection of ART and ATN. This suggests the presence of functional interactions between the pathways underlying chloroquine and artemisinin resistance or that resistance to cloroquine genetically creates genetic conditions which facilitate the appearance of ART and ATN resistance. These questions have significant relevance to the practical lifetime of a drug in areas where resistance to other drugs (or to chloroquine specifically) is prevalent. The inspection of mutation or mutations underlying resistance to ART and ATN, will be described in Chapter IV (known or suspected mutations) and in Chapter V using highthroughput comparative genomic studies based on genome-wide approaches, Linkage Group Selection (LGS) (new mutations) [Culleton R et al. 2005]. 136 137 RESULTS CHAPTER IV ANALYSIS OF THE PUTATIVE GENETIC MODULATORS FOR ARTEMISININ AND ARTESUNATE RESISTANCE 144 4.2. Estimation of gene copy numbers of the pcmdr1, pctctp and pcatp6 genes, in the selected mutant clones. Besides gene mutations, changes in copy numbers may also be responsible for drug resistance. Therefore, gene copy number of pcmdr1, pctctp and pcatp6 on the previously selected and cloned artemisinin and artesunate mutants in P. chabaudi (AS-ART and ASATN respectively) were studied. At this stage the gene pccg10 was not analysed for gene copy number change due to the fact that there were no references found in the literature that account for changes in the gene copy number of either pfcrt or its P. chabaudi homologue pccg10. The gene copy number can be analysed by various methods, the method of using real-time quantitative PCR (RTQ-PCR) has been extensively validated for the use in malaria parasites. We found that the most robust method for copy number determination by real-time PCR is the comparative Ct (2 -∆∆Ct ) method [Livak KJ et al. 2001]. While requiring an endogenous control and a calibrator, meaning a gene from which previous knowledge is needed for the exact gene number (for example msp-1 gene that is known for sure to be a single copy gene in P. chabaudi) and a sample that is use as a control for that gene, it differs from relative standard method by relying on equal PCR efficiencies with the target and the endogenous control genes. The 2 -∆∆Ct method is described in detail by Livak and colleagues [Livak KJ et al. 2001]. Briefly for the ∆∆Ct calculation to be valid, the efficiency of the amplification of target and reference gene must be approximately equal. The control gene was msp-1 and the target genes were mdr1, tctp and atp6 genes. The average Ct was calculated for both control and target genes and the ∆Ct (Ct target gene – Ct msp-1 ) was determined. Different plots of the log DNA dilution versus ∆Ct were made and whenever the slope was close to zero the efficiencies of the target and reference genes were similar, the ∆∆Ct was calculated for the relative quantification of the target gene; ∆∆Ct= (Ct target gene - Ct msp-1 ) α - (Ct target gene - Ct msp-1 ) β , where α = ART or ATN resistant sample and β = AS-30CQ or AS-15CQ samples respectively. After the method was validated the results for each sample were expressed in N-fold numbers in α gene copy number with normalization to the msp-1gene copy number according to the equation: gene number of target = 2 -∆∆Ct [Livak KJ et al. 2001]. After optimisation of all the RTQ-PCR conditions we aimed to investigate if ART and ATN resistance could be related to changes in the gene copy number of mdr1, tctp and atp6 genes. For this the N-fold gene number was evaluated and mdr1, tctp and atp6 genes were normalised to msp-1 gene, in artemisinin resistant parasites P. chabaudi AS-ART relative to 145 artemisinin sensitive P. chabaudi AS-30CQ from which this parasite line was obtained and artesunate resistant P. chabaudi AS-ATN relative to artesunate sensitive P. chabaudi AS15CQ from which this parasite line was obtained. Three independent experiments were carried out; the mean N-fold values of the experiments are presented in Figure 27. There were no changes in the gene copy number of mdr1, tctp and atp6 genes detected (Figure 27), meaning that during the selection for artemisinin and artesunate resistance the genes pcmdr1, pctctp and pcatp6 did not suffer any gene copy number change. Figure 27 – Relative differences (N-fold) in gene copy number between artemisinin (AS-ART) and artesunate (AS-ATN) resistant parasites and their sensitive progenitors, AS (30CQ) and AS (15CQ) respectively. The y-axis represents the mean N-fold of gene number (grey bars) and standard deviations (vertical lines) at 95% confidence interval of each of the genes under study normalised against msp-1, generated after three independent assays. From: Afonso A et al 2006. 0 0,5 1 1,5 2 AS-ART AS-ATN AS-ART AS-ATN AS-ART AS-ATN mdr1 tctp atp6 0 0,5 1 1,5 2 0 0,5 1 1,5 2 AS-ART AS-ATN AS-ART AS-ATN AS-ART AS-ATN mdr1 tctp atp6 146 4.3. Discussion 4.3.1. Summary The previously described putative genetic modulators for artemisinin and artesunate drug resistance, pcmdr1, pcg10, pctctp and pcatp6 that are P. chabaudi gene homologues of P. falciparum mdr1, crt, tctp and atp6, respectively, were fully isolated and analysed for their coding regions. In the case of the gene atp6 gene an area of 4kb upstream and 1 kb downstream was also analysed, in both the resistant clone lines, AS-ART and AS-ATN in relation to their sensitive progenitors AS-30CQ and AS-15CQ respectively and found to be not mutated. Collectively, the above data allowed us to conclude that in our model P. chabaudi the pcmdr1, pcg10, pctctp and pcatp6 genes are not involved in artemisinin or artesunate resistance by gene point mutation. Besides looking for point mutations, the genes pcmdr1, pctctp and pcatp6 were also investigated for their involvement in artemisinin and artesunate resistance by inspecting putative changes in gene copy number. This allowed us to conclude that the phenotype of artemisinin and artesunate resistance in our model, P. chabaudi, is also not associated with an increase in the gene copy number of the same genes. 147 4.3.2. General Discussion This study has shown that no changes in nucleotide sequence or copy number for pcmdr1, cg10, tctp or atp6 genes were found in the artemisinin or artesunate resistant parasites when compared to their sensitive progenitors. We consider, however, that other genetic mechanisms such as protein turnover and/or posttranslational modifications of the gene products may account for the involvement of these genes in the resistance phenotype. However the investigation of these putative mechanisms was not contemplated in this study. In malaria, the genetic mechanisms underlying drug resistance have been extensively studied for most antimalarials, but are not fully understood, except for resistance to pyrimethamine. Resistance to this drug has been to shown to be conferred by cumulative single nucleotide mutations in the dhfr gene [Sirawaraporn W et al. 1997] in P. falciparum and in all rodent malaria models analyzed. Artemisinin has been used in the field for centuries without resistance ever being registered, so one might expect a complex mechanism for resistance, not as simple as that for pyrimethamine. In our study in a rodent malaria model, we selected for high levels of drug resistance, which suggests the involvement of more than one gene. The evidences supporting the role for pfatpase6 protein in artemisinin sensitivity in P. falciparum arise from the interaction of artemisinin in an ex vivo heterologous system (Xenopus oocytes) [Eckstein-Ludwig U et al. 2003] which may reflect a histological, physiological and biochemistry point of view which may be difficult to interpret in light of the human malaria parasite. Evidence for the involvement of this gene in modulating artemisinin susceptibility also came from the fact that recent evidence indicates that mutations in the pfatpase6 gene may correlate with varying degrees of in vitro responses of P. falciparum to artemisinin derivatives [Jambou R et al. 2005]. In this case the pfatpase6 protein S769N, A623E and E431K polymorphisms were associated with an increased mean in the IC50 for artemisinins [Jambou R. et al. 2005]. The importance of these findings is unclear however, since: i) the correspondence between reduced sensitivity of the field isolates to artemisinins and the presence of the different polymorphisms was incomplete and ii) there was a regionspecific association of these polymorphisms with varying degreees of susceptibility correlating highly (but not completely) with the S769N mutation in French Guiana but not in Senegal or Cambodia. In addition, later studies found no association between mutations in the pfatpase6 gene and the sensitivity of field isolates to artemisinins from Tanzania [Mugittu K et al. 2007] and São Tomé and Principe [Ferreira ID et al. 2007]. 148 Interestingly, it may be possible that another gene closely link to the atp6 gene may be involved in the resistance to this drug, which could explain the strong but incomplete association observed in French Guiana [Jambou R. et al. 2005]. In actual fact, a similar scenario has already been described before with the chloroquine resistance determinant in P. falciparum which had been mapped by linkage analysis to a segment on chromosome 7 where two candidate genes cg1 and cg2 with complex polymorphisms initially linked to the chloroquine resistance phenotype where found [Duraisingh MT et al. 2000a, Fidock DA et al. 2000a]. More detailed analysis within that region later revealed a different, but closely linked gene, pfcrt, to be the major determinant of chloroquine resistance. Finally, it is relevant to mention that, if pfatpase6 turns out to be the major modulator of artemisinin responses in P. falciparum, it is conceivable that the rodent malaria P. chabaudi may reveal alternative mechanisms of resistance to those of P. falciparum, as is the case with chloroquine resistance [Hunt P et al. 2004]. 149 RESULTS CHAPTER V EXPERIMENTS USING LINKAGE GROUP SELECTION AS AN ATTEMPT TO IDENTIFY THE LOCUS OR LOCI INVOLVED IN ARTEMISININ AND ARTESUNATE RESISTANCE 150 151 5.1 Introduction As described in the previous chapter (Results – Chapter IV) sequence analysis of genes with potential involvement in modulating parasite responses to artemisinin derivatives, was carried out for the P. chabaudi orthologues of pfatp6 [Eckstein-Ludwig U et al. 2003; Jambou R et al. 2005; Uhlemann AC et al. 2005], pfcrt [Sidhu AB et al. 2002], pfmdr1 [Ferrer-Rodríguez I et al. 2004; Price RN et al. 2004; Reed MB et al. 2000; Sidhu AB et al. 2005], and pftctp [Bhisutthibhan J et al. 1998; Walker DJ et al. 2000]. Sequencing of AS-ART and AS-ATN and their progenitors showed that there were no mutations or copy number changes in these genes (data published Afonso A et al. 2006). This chapter describes the application of Linkage Group Selection (LGS) as an approach to identify the genetic locus or loci involved in artemisinin and artesunate resistance in P. chabaudi. Some of the results described below have been accepted for publication [Hunt P et al. in press at the Molecular Microbiology]. 152 5.2 Production of cross progeny The procedure to obtain a cross progeny have been already described (See Chapter II - Materials and Methods). Three independent genetic crosses were obtained using AS-ART and AJ and two independent genetic crosses were obtained using AS-ATN and AJ. One of the conditions for a correct and useful utilization of LGS for identifying genetic locus involved in a particular phenotypic trait, is that large numbers of recombinant progeny clones are present both in the progeny prior to the selection, and post selection. For that reason and in order to increase the number of recombinants, the uncloned progeny from these crosses (called “unpassaged”) were pooled in equal proportions (equal number of parasites) and passaged through the two treated and untreated groups. See Figure 28 for details on the experimental groups. For further analysis pooled crosses were used. Those pooled crosses were named super crosses. The three genetic crosses obtained between AS-ART and AJ were pooled together in a super cross that for clarity was named AS-ART x AJ. The two genetic crosses obtained between AS-ATN and AJ were polled together in a super cross that for clarity was named ASATN x AJ. Each cross was not analysed individually during this project. 153 5.3 Selection of cross progeny 5.3.1 General procedure When the sporozoites-induced infections (so called “unpassaged”) reached parasitaemias of between 10%–15%, the parasites were harvested for AFLP analysis (providing a reference point for markers analysed in the subsequent treatment groups), pooled, and inoculated (each mouse in each group received 1 x 10 7 parasites) into two groups of mice, one drug treated with either artemisinin or artesunate (so called “treated”) and the other left untreated (so called “untreated”). See Figure 28 for details. Figure 28 - Schematic representation of the selection of the pooled cross progeny (LGS) experiment using previously generated individual crosses between AJ and AS-ART For the pooled cross AS-ATN x AJ only two individual crosses were used. Adapted from Culleton R, 2005 with kind permission from Dr. Richard Culleton. Genetic cross 1 Genetic cross 2 Genetic cross 3 Unpassaged Super cross pooled in equal proportion of parasite numbers from individual crosses Treated Untreated