QUANTIFYING THE INTENSITY OF PERMETHRIN INSECTICIDE RESISTANCE IN ANOPHELES MOSQUITOES IN WESTERN KENYA. OMONDI SELINE AWUOR, BSc I56/74247/2014

Transcription

1 QUANTIFYING THE INTENSITY OF PERMETHRIN INSECTICIDE RESISTANCE IN ANOPHELES MOSQUITOES IN WESTERN KENYA OMONDI SELINE AWUOR, BSc I56/74247/2014 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE AWARD OF THE DEGREE OF MASTER OF SCIENCE IN MEDICAL AND VETERINARY ENTOMOLOGY OF THE UNIVERSITY OF NAIROBI 2017 i

2 DECLARATION I declare that this thesis is my original work and has not been presented for the award of a degree in any other university. CANDIDATE SIGNATURE: DATE: Ms. SELINE AWUOR OMONDI, Bsc SUPERVISORS SIGNATURE: DATE: WOLFGANG RICHARD MUKABANA, PhD Associate Professor, School of Biological Sciences, University of Nairobi, Nairobi, Kenya SIGNATURE: DATE: DR. NABIE BAYOH Technical Advisor, Entomology Section, Kenya Medical Research Institute/Centers for Disease Control and Prevention ii

3 ACKNOWLEDGEMENTS First and foremost thanks to the almighty God for giving me the strength, knowledge, ability and opportunity to undertake this research. Secondly, tremendous thanks to my father Amos Omondi, you gave me support in this work and made sure I went to the field even when funds were unavailable. This achievement would not have been possible without you. I extend my appreciation to my supervisors; Dr. Nabie Bayoh and Prof. Wolfgang Richard Mukabana for giving me advice and valued mentoring at all times. I take pride in acknowledging Dr. Eric Ochomo, he was always willing to help, give guidance and equip me with knowledge concerning insecticide resistance. I cannot fully express my gratitude to Prof. Henry Ouma and Dr. Peter Ouma for their superb guidance and also for introducing me to the world of entomology. Also Special thanks to my friend Margaret Muchoki, for sharing her knowledge in statistics. I would like to acknowledge the institutions that supported my training and research; University of Nairobi and KEMRI/CDC. Lastly, I owe my deepest appreciation to Amito, Brigid and Esalimba for taking me through the field and laboratory work. I was honored to have you as my teachers. God bless you all. iii

4 DEDICATION To my dad, Amos Omondi. iv

5 TABLE OF CONTENTS DECLARATION... ii ACKNOWLEDGEMENTS...iii DEDICATION... iv TABLE OF CONTENTS... v ABBREVIATIONS...viii LIST OF TABLES... ix LIST OF FIGURES... x ABSTRACT... xi CHAPTER 1.0: INTRODUCTION Justification and significance of the research Main objective Specific objectives Statistical hypothesis Null hypothesis Alternative hypothesis... 4 CHAPTER 2.0: LITERATURE REVIEW Mosquito biodiversity Diseases transmitted by mosquitoes Anopheles mosquito species Malaria parasites and their distribution Control measures for malaria Control of malaria parasites v

6 2.3.2 Control of Malaria Vectors Assays for testing insecticide resistance WHO Susceptibility test CDC bottle assay Cone bioassay CHAPTER 3.0: MATERIALS AND METHODS Study site Mosquito sampling and rearing Status of permethrin resistance in malaria vectors in western Kenya Intensity of Permethrin Resistance in malaria vectors in western Kenya Preparation of resistance diagnostic treatments Exposing mosquitoes to resistance diagnostic treatments Effect of intensity of Permethrin Resistance on efficacy of insecticide treated nets Performance of Cone Bioassays Identification of Anopheles gambiae complex mosquitoes into species Extraction of mosquito genomic DNA PCR amplification Separation of PCR amplified products by gel electrophoresis Visualization of separated, PCR-amplified products Data Analysis CHAPTER 4.0: RESULTS Status of permethrin resistance in malaria vectors in western Kenya Intensity of permethrin resistance in malaria vectors in western Kenya vi

7 4.3 Effect of intensity of permethrin resistance on efficacy of insecticide treated nets Molecular assays CHAPTER 5.0: DISCUSSION CONCLUSION RECOMMENDATIONS REFERENCES APPENDICES... 1 APPENDIX 1: WHO susceptibility test recording form... 1 APPENDIX 2: CDC bottle bioassay data recording form... 2 APPENDIX 3: Cone bioassay data recording form... 3 vii

8 ABBREVIATIONS ACT Atemisinin-based Combination Therapies CDC Centers for Disease Control and Prevention CGHR Centre for Global Health Research DDT Dichlorodiphenyltrichloroethane DHIS District Health Information System GST Glutathione-S-transferase IRS Indoor Residual Spraying KD Knock down Kdr Knock down resistance KDT 50 Knock Down Time 50 (The time for 50% knock down) KEMRI Kenya Medical Research Institute KMIS Kenya Malaria Indicator Survey LLINs Long Lasting Insecticidal Nets PBO Piperonyl butoxide PCR Polymerase Chain Reaction PMI Presidents Malaria Initiative s.l. sensu lato s.s. sensu stricto WHO World Health Organization viii

9 LIST OF TABLES Table 1: PCR master mixture Table 2: WHO susceptibility assay of Anopheles gambiae s.l to 0.75% permethrin Table 3: Knock down (KD) and mortality rates of Anopheles mosquitoes exposed to the three LLIN brands Table 4: Species identification by study site ix

10 LIST OF FIGURES Figure 1: Global distribution of Anopheles mosquitoes (adopted from Sinka et al., 2012)... 7 Figure 2: Global distribution of malaria (source: WHO World malaria report, 2014)... 8 Figure 3: Endemicity of Malaria in Kenya (source: Kenya Malaria Indicator Survey 2015). 10 Figure 4: The WHO insecticide susceptibility assay for testing insecticide resistance Figure 5: The CDC bottle bioassay for testing insecticide resistance Figure 6: Anopheles mosquitoes exposed to LLIN using cone bioassay Figure 7: Map of Kenya showing the study sub-counties of Bondo, Nyando, Rachuonyo and Teso Figure 8: Mosquito larval sampling in a pool of water in Kaliwa, Teso sub-county, western Kenya Figure 9: Sorting mosquito larvae in Kamenya, Rachuonyo sub-county, before transportation to the KEMRI CGHR centre Figure 10: KDT 50 values at 95% CI for female Anopheles mosquitoes exposed to 1, 2, 5 and 10 permethrin concentrations in the CDC bottle bioassay x

11 ABSTRACT The development and spread of resistance among local mosquito vectors to the major classes of insecticides used in Long Lasting Insecticidal Nets (LLINs) and Indoor Residual Spraying (IRS) poses a major challenge to malaria vector control programs. The main methods of evaluating insecticide resistance in malaria vector include the WHO tube bioassay and CDC bottle assays. The main weakness in these methods is that they measure resistance at fixed standardized insecticide concentrations and not variable insecticide doses to determine intensity of resistance. However the CDC bottle assay using different insecticide dosages has proved applicable in ascertaining the intensity of resistance. This study determined the status and intensity of permethrin resistance and investigated the efficacy of commonly used LLINs (PermaNet 2.0, PermaNet 3.0 and Olyset ) against 3-5 day old female Anopheles mosquitoes from four sub-counties; Teso, Bondo, Rachuonyo and Nyando in Western Kenya. Two sentinel sites were selected per sub-county; Akiriamasi and kaliwa in Teso, Barkanyango and Omia Mwalo in Bondo, Kobuya and Kamenya in Rachuonyo and lastly Kochogo and Ahero in Nyando. The mosquitoes were exposed to three different assays; WHO susceptibility test, CDC bottle assay and cone bioassay. For the intensity of insecticide resistance the mosquitoes were exposed to 1(21.5µg/ml), 2(43µg/ml), 5 (107.5µg/ml) and 10(215µg/ml) doses of permethrin using CDC bottle assays. Mortality after 24h was recorded for WHO assays; time to knockdown was monitored for the bottle assays while both mortality and knockdown was used for the cone assay. All the mosquito samples were identified to species level using standard polymerase chain reaction. It is only in Kaliwa, Teso were An. gambiae s.s was the major species while the rest were An. arabiensis. Resistance to permethrin was confirmed in Teso, Bondo, Rachuonyo and one site from Nyando while there was likelihood of resistance in Kochogo, Nyando. Intensity of permethrin resistance was highest in Barkanyango, Bondo with 84% knockdown at the 30 minutes diagnostic time when exposed to the 10 dose. Mortality for the nets was (0-39) %, (12-88) % and (26-89) % for Olyset, PermaNet 2.0 and PermaNet 3.0 respectively. The efficacy of the nets was greatly reduced in Bondo and Teso. Results from this study shows that there was confirmed resistance in all the sites: However, intensity assays were able to differentiate Bondo and Teso as the sites with the highest levels of resistance, which coincidentally were the two with reduced net efficacy. There was a reduced efficacy of nets in areas with high resistance portraying that at certain intensities of resistance, vector control using LLINs may be compromised. Therefore it is necessary to incorporate intensity assays in order to determine the extent of threat that resistance poses to malaria control. xi

12 CHAPTER 1.0: INTRODUCTION Malaria is a mosquito borne disease transmitted by female Anopheles mosquito species and one of the leading causes of deaths worldwide (WHO, 2015a). The principle vectors of malaria in Kenya are Anopheles gambiae, An. arabiensis, and An. funestus. The disease accounts for the majority of morbidity and mortality in Kenya with a 16.4% of outpatient attendance (DHIS2, 2014). Illnesses and deaths due to malaria are severe among children below the age of five years and pregnant women. There has been a global decline of malaria cases in the last ten years but the disease is still a threat (WHO, 2015a). Much of the recent decrease in the global malaria burden has been achieved through scale up of vector control interventions mostly via the use of insecticides in indoor residual spraying (IRS) and in long lasting insecticidal nets (LLINs) and other materials (WHO, 2006). The most recommended insecticides are pyrethroid based. This is because they offer several advantages in terms of cost, safety and duration of residual action as compared to the other classes of insecticides (WHO, 2011). Despite the national and global efforts the disease remains a burden in sub Saharan Africa. The use of just one insecticide class, which is the pyrethroids has given rise to new concerns about problems of resistance to insecticides in malaria vectors. Wide scale use of insecticide based malaria control strategies over the past decade has been associated with the development of resistance in several important malaria vector species. Pyrethroid resistance is most common in sub Saharan Africa and has been reported in 27 countries, including Kenya (WHO, 2015a). This drawback threatens the robustness of pyrethroid based interventions. The only action is therefore to adopt a proactive approach and modify 1

13 contemporary practices so as to delay the spread of resistance and preserve the effectiveness of current insecticides at least until new ones are developed. To do this, malaria regions should regularly monitor the presence of resistance as well as the strength of resistance that may affect the efficacy of the insecticide based vector control interventions in reducing the burden of malaria. Phenotypic resistance can be evaluated using the WHO susceptibility test and the CDC bottle assays (WHO 2013, Brogdon and Chan 2010). Up until the time this study was conducted, the WHO susceptibility test only provided information on whether a particular mosquito population is susceptible or resistant at the 90% level of mortality and did not provide much detail on varying levels of intensity of resistance. For example, a population with 10% mortality to an insecticide and another with 89% mortality to the same insecticide were both classified as resistant, despite the fact that, programmatically, the two levels were bound to have different levels of impact to malaria control. For instance western Kenya has been reported to have high frequencies of pyrethroid resistance (Ochomo et al., 2013) but the intensities of this resistance and whether or not it impacts on the burden is not known. Majority of LLINs in western Kenya are impregnated with permethrin or deltamethrin. Analysis performed on mortality data for deltamethrin and permethrin have indicated the two are positively correlated (Ochomo et al., 2013) and therefore only one insecticide was used for this study. In this study, a variety of quantitative bioassays were performed to determine the intensity of permethrin resistance on field collected An. gambiae sensu lato. The samples were collected in two sentinel sites from each one of four sub-counties namely Bondo, Nyando, Rachuonyo and Teso in western Kenya. 2

14 1.2 Justification and significance of the research Insecticide resistance is a major challenge to the current vector control strategies involving use of Long Lasting Insecticidal nets and Indoor Residual Spraying. Evaluation of insecticide resistance in malaria vector populations follows WHO protocols which involve the use of susceptibility tube bioassays with papers coated with fixed dose of insecticides. While the protocols provide useful information of prevalence of insecticide resistance in any vector population, it does not quantify the intensity of insecticide resistance. Furthermore, the concentration of insecticide used in resistance monitoring has no relationship with the quantities used in field application. Consequently, there is a high likelihood that the use of status (susceptible or resistance) of insecticide resistance in resistance monitoring does not provide adequate information on the extent of threat that resistance poses to malaria control, with a potential failure of the control measures. It is therefore important to redefine the way in which resistance is measured in the field to enable researchers define thresholds of resistance above which insecticide based vector control interventions may not be efficacious. Thus the study evaluated quantitative bioassays to compare resistance status indicated by WHO tube assays, the CDC bottle assays depicting the highest concentration of insecticide that can kill resistant field populations of Anopheles mosquitoes and the impact of this resistance on commonly used LLINs in western Kenya. This cleared the way for calibration of strength of resistance that can be used to monitor impact of resistance on vector control interventions especially LLINs. 3

15 1.3 Main objective To investigate the relationship between status and intensity of permethrin resistance along with their effect to the commonly used LLINs in malaria vectors in western Kenya Specific objectives 1. To determine the status of permethrin resistance in malaria vectors in western Kenya 2. To investigate the intensity of permethrin resistance in malaria vectors in western Kenya 3. To evaluate the effect of permethrin resistance on efficacy of insecticide treated nets against malaria vectors in western Kenya. 1.4 Statistical hypothesis Null hypothesis 1. There is no difference in the status of permethrin resistance in the four sub-counties in western Kenya. 2. There is no difference in the intensity of permethrin resistance in malaria vectors in western Kenya 3. There is no difference in the efficacy of PermaNet 2.0, PermaNet 3.0 and Olyset long lasting insecticidal nets against malaria vectors in western Kenya Alternative hypothesis 1. There is a difference in the status of permethrin resistance in the four sub-counties in western Kenya. 2. There is a difference the intensity of permethrin resistance in malaria vectors in western Kenya 3. There is a difference in the efficacy of PermaNet 2.0, PermaNet 3.0 and Olyset long lasting insecticidal nets against malaria vectors in western Kenya. 4

16 CHAPTER 2.0: LITERATURE REVIEW There are a vast number of insects that are responsible for diseases of public health importance. Some diseases for example malaria have been there since time immemorial while others have emerged as major problems of recent times. Vector borne diseases account for more than 17% of the infectious diseases and they are responsible for millions of deaths worldwide. Mosquitoes are the best known vectors but others include fleas, sandflies, ticks, triatomine bugs, tsetse flies etc. These are responsible for transmitting plague, leishmaniasis, tularemia, chagas disease and trypanosomiasis, respectively (Kloos and Berhane 2006). 2.1 Mosquito biodiversity Mosquitoes belong to the family Culicidae. Many members of this family are not only a nuisance to human beings, animals, reptiles, amphibians and birds due to their prominent blood sucking behavior but also are vectors of pathogens. Mosquitoes have a worldwide distribution and many species occur in the humid tropics and subtropics. Mosquitoes breed in water hence they are not found in the deserts and permanently frozen areas since these regions lack favorable aquatic breeding sites. Their species diversity varies greatly in the world with the majority found in the neotropical region, followed by oriental, then the Australian and finally the Afrotropical region. The lowest biodiversity of mosquitoes occur in the Nearctic region (Rueda, 2008) Diseases transmitted by mosquitoes The four major genera of the Culicidae family include Anopheles, Aedes, Culex and Mansonia. These act as either primary or secondary hosts for a number of parasites and viruses that cause diseases to vertebrates. The most dangerous disease is malaria which is transmitted by Anopheles species (Service 1993). Other diseases transmitted by anopheline mosquitoes include; lymphatic filariasis caused by Wuchereria bancrofti, Brugia malayi and 5

17 B. timori filarial worms. These worms are transmitted by the four major genera of the Culicidae family (Boakye et al., 2004; Zhang et al., 1991; Partono et al., 1977). Mosquitoes also transmit arboviral diseases e.g. dengue fever, yellow fever, west Nile fever, Japanese encephalitis, Eastern equine and others (Hubalek, 2008). Anopheles spp have also been found to transmit the nematode Dirofilaria immitis (dog heartworm) which depends on the vector for its maintenance and transmission (Ledesma and Harrington, 2011). Another genus, Toxorhynchites, constitutes giant mosquitoes that do not feed on blood as adult but their larvae feed on other larvae and are used as biological control agents Anopheles mosquito species Only Anopheles mosquitoes transmit malaria. Since malaria is the most devastating illness transmitted by mosquitoes, this characteristic renders the species dangerous. Anopheles mosquitoes are distributed all over the world and Sub Saharan Africa has a vast number of vector species (Figure 1). The main vectors of malaria in sub Saharan Africa belong to the An. gambiae complex and An. funestus complex groups. The An. gambiae complex is comprised of 9 sibling species, 7 of which are vectors (An. gambiae sensu strictu Gilles, An. arabiensis Patton, An. merus Donitz, An. melas Theobald, An. bwambae White, An. coluzzii Coetzee & Wilkerson and An. amharicus Hunt) and the 2 non vectors are (An. quadriannulatus Theobald and An. quadriannulatus species B) (coetzee et al., 2013). The An. funestus comprises of nine sibling species: An. funestus, An. vaneedenii Gillies & Coetzee, An. leesoni Evans, An. rivulorum Leeson, An. parensis Gillies, An. fuscivenosus Leeson, An. aruni Sobti, An. brucei Service and An. confuses Evans (Sinka et al., 2010). Among these An. gambiae sensu stictu, An. funestus and An. arabiensis are the primary vectors of malaria. 6

18 Figure 1: Global distribution of Anopheles mosquitoes (adopted from Sinka et al., 2012) 2.2 Malaria parasites and their distribution Malaria is the most detrimental parasitic infection caused by a parasite belonging to the genus Plasmodium and is a leading cause of deaths worldwide (WHO, 2015a). Transmission of malaria majorly occurs in the tropical and subtropical areas of the world (Service, 1997) (Figure 2). Transmission of the disease occurs when an infected female Anopheles mosquito bites a human host during the insects gonotrophic cycle (Klowden and Briegel, 1994). Several Plasmodium species namely P. falciparum, P. vivax, P. ovale and P. malariae cause malaria in humans. Plasmodium knowlesi has recently been incriminated to cause disease in man as well (Cox-Singh and Singh, 2008). Plasmodium falciparum is found in Africa, P. vivax in USA, Latin America and some parts of Africa, P. malariae in Sub Saharan Africa, South Asia, Indonesia, western pacific and in the Amazon, P. ovale is mainly found in West 7

19 Africa, the Philippines, Eastern Indonesia and Papua New Guinea (Lucas and Gilles, 1990). Anopheles gambie complex and An. funestus mosquitoes are known to transmit malaria in sub Saharan Africa (Coetzee, 2004; and Coetzee, 1993) where P. falciparum, the deadliest of all the Plasmodium parasites, is most prevalent and accounts for about 98% of all the malaria infection. Figure 2: Global distribution of malaria (source: WHO World malaria report, 2014) All the four Plasmodium species causing malaria occur in Kenya and they account for 16.4% of outpatient attendance (DHIS2, 2014). The high morbidity and mortalities are attributed to infections by P. falciparum. According to the latest reports by the World Health Organization (WHO) 438,000, malaria deaths and 214 million malaria cases were reported in 2015 majority being children aged under 5 years in Sub-Saharan Africa (WHO, 2015a). Globally malaria cases have decreased since the year 2000 but still it is a life threatening disease which takes the life of a child every 2 minutes (WHO, 2015a). Malaria places a 8

20 burden on health care systems and reduces productivity at homes and work places. The cost of treating the disease is high on households and health systems (Chima et al., 2003). The Kenyan government recognizes malaria as a health and socio- economic burden and gives it priority in that the goal of the Ministry of Health is to reduce the number of illness and deaths to 3% by 2030 (KMIS, 2011). The disease accounts for 18% of patients at risk and 27% of the population lives in endemic region (PMI 2016) (figure 3). Malaria risk in these regions is high since there is plenty of rainfall and favorable temperatures for the vectors survival. Temperatures are always high and water pools provide breeding areas for the vector. The vectors lifecycle is short and the survival rates are high due to suitable climatic conditions. These are high malaria transmission areas for all year long and prevalence rates range from 20% to 40% in the Lake Victoria regions and a prevalence of less than 5% for the Coastal regions. 9

21 Figure 3: Endemicity of Malaria in Kenya (source: Kenya Malaria Indicator Survey 2015) 10

22 2.3 Control measures for malaria Control measures are directed towards the disease (parasite) and the vectors. Malaria control involves combined approach to timely and effective diagnosis, and treatment for the parasite and preventive measures and killing of the vectors Control of malaria parasites Chemotherapy Malaria is treated using several drugs depending on the plasmodium species, drug resistance and how the disease has progressed. It is important to prevent the disease before it progresses from uncomplicated malaria to a more severe illness that leads to death. Patients with uncomplicated P. falciparum malaria are treated with artemisinin based combination therapies (ACT) while uncomplicated P. vivax malaria is treated with chloroquine. Pregnant women are given three doses of intermittent preventive treatment with sulfadoxinepyrimethamine (IPTp-SP) as a measure to prevent malaria in pregnancy. Non-pregnant women, men and children are treated with primaquine where P. ovale or P. vivax is present in order to prevent future relapse. All severe cases of malaria are treated with artesunate or artemether followed by full course of ACT (WHO, 2015a). Drugs have been administered in extreme degree of coverage but it is evident that the parasites have developed resistance to some of these drugs such that the expected results have not been achieved (Dondorp et al., 2009). Resistance to Chloroquine was discovered in 1993 by Warrel and this brought about the necessity to develop new chemicals to eliminate the parasites. Antemisinin based therapy has resulted in dramatic decline of the disease but these newly created drugs are expensive and not many people can afford them (Gelband et al., 2004) Vaccines 11

23 Malaria vaccine development is still on the clinical trial stages and not yet available for public use. Of the five species of Plasmodium that cause malaria in human beings, P. falciparum and P. ovale have received attention for development of vaccines (Schwartz et al., 2012). Mosquirex (RTS, S) is the vaccine candidate that is furthest in development and provides protection against Plasmodium falciparum. The vaccine triggers the immune system to prevent the parasite from infecting the liver tissues (Gosling and von Seidlein 2016). Another vaccine candidate, still on clinical trial is Plasmodium falciparum sporozoite (PfSPZ) vaccine which is highly protective against malaria infection three weeks after the immunization but the period that the protection lasts is still unknown (Ishizuka et al., 2016) Control of Malaria Vectors Control of mosquitoes remains an effective means of curbing malaria. Before DDT (dichlorodiphenyltrichloroethane) was discovered in the 1950s mosquito control was directed towards the immature stages (Smith et al., 2002). In the past malaria vector control relied heavily on community participation where detailed knowledge of the insect habitat led to environmental manipulation. Breeding sites of the mosquitoes were eliminated by drainage of water filled areas or by modification of the preferred habitat of the species i.e. clearing the river banks to improve the water flow (Mabaso et al., 2002). Mosquito larval control was majorly based on the use of larvivorous fish, petroleum oil and paris green which were applied on water bodies in order to kill the larvae (Mabaso et al., 2002). Other methods such as the use of plant extracts with some levels of success include; Neem (Azadirachta indica), basil (Ocimum basilicum) and eucalyptus (Corymbia citriodora) are used either as repellents, larvicides or insecticides (Queen et al., 2014). These methods were environmental friendly and had minimum threats to human beings. With time other 12

24 methods have been developed i. e the use of bacterial symbiont Wolbachia that confers resistance in mosquitoes to the human malaria parasite (Bian et al., 2013). Development of DDT made it possible to control not only mosquitoes but also other vectors of diseases at adult stage. During this time the idea of eliminating the vectors using DDT provided a promising possibility of eradicating the diseases. Other insecticides have been discovered and developed since then due to the life threatening secondary effects of DDT but surprisingly the world still faces a major threat from malaria as it did in the 1950s Insecticides This is the most effective measure in controlling malaria vectors. The use of insecticide based vector control interventions such as LLINs and IRS are a common and widespread malaria control strategy in areas where the disease is rampant (WHO, 2015a). Before and after biting a human host endophagic mosquitoes rest on the nets, walls, ceiling or other places in the house (Githeko et al., 1996). If the surfaces have been sprayed by residual insecticides the vector picks up a lethal dose which in turn kills the mosquito before the parasite matures in its system. Long lasting insecticidal nets on the hand not only kill the vectors but also play a major role in being a physical barrier preventing human-vector contact. Both IRS and LLINs reduces the mosquito densities therefore reducing malaria transmission. The malaria control interventions have contributed tremendous to the decline in the malaria cases between 2001 and This decline was attributed majorly to the use of LLINs, IRS and ACT accounting for 69%, 10% and 21% respectively (WHO, 2015a). Both IRS and LLINs rely on the use of insecticides but where IRS has a vast number of insecticides to choose from, LLINs can only be impregnated with insecticides belonging to a single class. 13

25 The classes of insecticides used in mosquito control are organochlorines, organophosphates, carbamates and pyrethroids (N guessan et al., 2007). These insecticides are both used indoors for public health in control of vectors and outdoors for agricultural purposes. Organochlorines and the pyrethroid insecticides act by altering the sodium ion channels while organophosphates and carbamates acts as acetyl cholinesterase (AchE) inhibitors thereby affecting the peripheral and central nervous system of the insects (Davies et al., 2007). The pyrethroids are widely used in Africa and are a major class of neurotoxic class while the organochlorine, Dichlorodiphenyltrichloroethane (DDT) is still used in some countries in Africa like Tanzania in IRS programs (Nkya et al., 2014) despite its adverse effects on life. The World Health Organization recommends the use of synthetic pyrethroids for LLINs since they have low mammalian toxicity (Zaim et al., 2000). There are two types of pyrethroids; type I and type II. Type I are naturally occurring insecticidal esters of chryysanthemic acid and is comprised of permethrin and bifentrin while type II are esters of pyrethric acid (Soderland, 2012). The type II pyrethroid consists majority of the insectides; deltamethrin, alpha-cypermethrin, lambdacyhalothrin and cyfluthrin (Thatheyus and Selvam 2013). Except for permethrin the rest are used in indoor residual spraying programs while alpha-cypermethrin, deltamethrin and permethrin are used in long lasting insecticidal nets. The pyrethroids affect the peripheral and central nervous system of the insects (Soderland, 2008). This is the same effect compared to the DDT but more pronounced. They stimulate the nerve cells by delaying the closure of voltage-gated sodium ion channel leading to repeated nerve firing and eventually paralysis (Davies et al., 2007). From the year 2001 to 2013 expansion of malaria vector control interventions contributed to a 47% decline in malaria mortality globally (WHO, 2013a). Despite all the efforts malaria still remains endemic and the burden is heaviest in Africa (WHO, 2015b) since these control 14

26 efforts are now being threatened by the widespread emergence of insecticide resistance (WHO, 2013a) Insecticide resistance among the malaria vectors Mosquitoes have been subjected to multiple insecticides since the synthetic organic insecticides were discovered. These insecticides have been and are still used in almost every part of the world. Resistance has developed and it threatens the robustness of the vector control techniques available. Agricultural activities and industrial pollutants have a high impact on control of the vectors, in that pesticides used against crop pests and the industrial waste products that possess insecticidal properties have the same target sites as those of insecticides targeting the mosquitoes (Blakey et al., 2013). In Africa most countries have reported increased resistance in mosquitoes collected from cotton cultivation sites and vegetable production areas (Reid and McKenzie, 2016). Reduced pyrethroid susceptibility have been recorded in west, central and East Africa and over 60 countries have reported resistance to at least one insecticide and some to all insecticide classes) (PMI 2015). The organophosphates; malathion, fenitrothion and pirimiphos-methyl so far are still promising and give high mortalities (Hammad et al., 2015, Riveron et al., 2015,Brooke et al., 2015) but resistance to these insecticides has been recorded in some areas (Sangba et al., 2016). Carbamates are less effective compared to the organophosphates, reports of resistance to bendiocarb and propoxur which are the only insecticides belonging to this group, have been reported (Sande et al., 2015, Riveron et al., 2015). Insecticide resistance is based on improved enzymatic detoxification and alteration of the insecticide target sites which are responsible for insecticide sensitivity (Hemingway et al., 2004). Improved enzymatic detoxification has been linked to three classes of enzymes, 15

27 monooxygenases, glutathione S-transferases (GSTs) and non-specific esterases. Anopheles gambiae possesses cytochrome P450 which is a member of the monooxygenase class of detoxifying enzymes. Resistance to pyrethroids relies mainly on the knockdown resistance target site mutation (kdr) and metabolic resistance but other mechanisms maybe involved i.e. cuticle alteration (Nkya et al., 2013). Dichlorodiphenyltrichloroethane resistance is conferred by elevated levels of detoxifying enzymes and also kdr mutation while carbamate resistance is caused by acetylcholinesterase mutation and detoxification (Mertinez-Torres et al., 1998; Ranson et al., 2000; Brooke et al., 2001). The recurrent use of these insecticides is the main cause of resistance in mosquito populations (Protopopoff et al., 2008). Due to this resistance to insecticides, controlling mosquitoes has become a great challenge and monitoring insecticide resistance has become a necessary and compulsory component of vector control interventions (Ranson et al., 2011). There are two main assays used in measuring insecticide resistance in adult mosquitoes, the WHO tube bioassay and the CDC bottle bioassay. 2.4 Assays for testing insecticide resistance There is little guidance on what the magnitude of insecticide resistance effect would be on long lasting insecticidal nets and indoor residual spraying. The WHO susceptibility assay and CDC bottle bioassays are used on large-scale in areas where malaria is endemic. The assays are useful in surveillance of insecticide resistance WHO Susceptibility test 16

28 The World Health Organization has been providing support to countries in monitoring and managing insecticide resistance. As part of this role it has developed standard test procedures and operational standards for detecting and monitoring insecticide resistance for several disease vectors including mosquitoes. The standard test for the mosquitoes is known as the WHO susceptibility test (WHO, 2013b). This is a simple direct exposure test that distinguishes between susceptibility and resistance to insecticides in adult mosquitoes. The test uses papers coated with discriminatory doses of insecticides i.e. 0.75% and 0.05% for permethrin and deltamethrin, respectively. Data are reported as percentage mortality and less than 90% mortality is used to define resistance, between 90% and 97% mortality defines the likelihood of resistance and 98% mortality and above defines susceptibility. The test is used for field surveillance and laboratory mosquito population monitoring. A representation of the assay is shown on Figure 4. Figure 4: The WHO insecticide susceptibility assay for testing insecticide resistance 17

29 2.4.2 CDC bottle assay An alternative method known as the CDC bottle assay was developed by the USA Centers for Disease Control and Prevention (CDC). It is a surveillance tool for detecting insecticide resistance in vector populations and is designed to determine if a particular formulation of an insecticide is able to control a vector at a specific location at a given time. The assay is cost effective in that one does not have to order specialized equipment as in the case of WHO susceptibility test since it uses 250ml glass bottles that are reusable (Brogdon and Chan, 2010). The bottles are coated with the insecticides of different concentrations in ascending order; 1, 2, 5 and 10 (21.5µg/ml, 43µg/ml, 107.5µg/ml and 215µg/ml respectively). Compared to the WHO susceptibility test, the CDC bottle assays can test for resistance at different insecticide concentrations and can allow the use of a synergist to analyze metabolic resistance (Owusu et al., 2015). The only disadvantage is that the CDC bottle assay is prone to contaminations i.e moisture build up inside the bottle and accidents where brakeages might occur during preparation of the bottles. An illustration of the CDC bottle bioassay is shown on figure 5. 18

30 Figure 5: The CDC bottle bioassay for testing insecticide resistance Cone bioassay Long lasting insecticidal nets (LLINs) have played a major role in combating malaria in large scale settings ever since the WHO recommended the universal coverage in Cone bioassays are used to assess the response of field mosquitoes to the field doses and formulations of insecticides being used either in nets or on walls. This assay allows for tests on the efficacy of the LLINs. Since the WHO susceptibility test uses insecticide concentrations which do not have any reference to the concentrations used on the nets, it is therefore necessary to monitor resistance and the nets performance in general (Bagi et al., 2015). An illustration of the assay is shown in figure 6. 19

31 Figure 6: Anopheles mosquitoes exposed to LLIN using cone bioassay. 20

32 CHAPTER 3.0: MATERIALS AND METHODS 3.1 Study site This was a stratified sampling design study conducted in four sub-counties in western Kenya namely Bondo (0 14 N E), Rachuonyo (0 30 S E), Nyando (0 11 S E) and Teso (0 43 N E). These are malaria endemic regions in western Kenya. Two sublocations were selected per sub-county for the study in terms of vector abundance: Akiriamasi and Kaliwa in Teso sub-county, Kamenya and Kobuya in Rachuonyo sub-county, Kochogo and Ahero in Nyando sub-county and finally Barkanyango and Omia Mwalo in Bondo sub-county. Malaria prevalence in these areas is high and perennial with seasonal peaks between April to July and November to December (Hamel et al., 2011). Bondo sub-county is in Siaya County and is located 50 kilometers west of Kisumu town, capital of the former Nyanza province. The sub-county has a population of 157,522 persons (2009 census). Rachuonyo sub-county is part of Homa Bay County and has a population of 382,711 (2009 census). Nyando sub-county is part of Kisumu County and is an agricultural area where rice farming is the major economic activity. It has a total population of 350,353 (2009 census) and is located 20 km east of Kisumu town. Nyando River flows through it and helps in irrigating the rice fields. Teso sub-county includes Malaba, a border town between Kenya and Uganda and is part of Busia County. The people of Teso are successful farmers in western Kenya and tobacco is the main cash crop in the area. Teso sub-county has a population of 255,871 (2009 census). These sub-counties have benefited either from Indoor Residual Spraying and Insecticidal Nets distribution for the past years (Hamel et al., 2011). Nyando and Rachuonyo have benefitted from both IRS and LLINs whereas Teso and Bondo only LLINs have been deployed. Mass distribution of LLINs was done in the year 2006 and these were with either Olyset (Sumutomo chemicals) or Permanet 2.0 (Vestergaard Frandsen) nets in Kenya (Gimnig et al., 2016). In 2008 and 2009 IRS first and second rounds were done in Rachuonyo using Lamda cyhalothrin (ICON CS,Syngenta) and Alphacyhalothrin 21

33 (Fendona,BASF) respectively(gimnig et al., 2016). In 2010 both Nyando and Rachuonyo were targeted for IRS and this was done using deltamethrin (K-Othrine, Bayer), a repeat was done in 2011 with lambda cyhalothrin(icon CS,Syngenta) in the two sub-counties (Kleinschmidt et al., 2015). In the last couple of years pyrethroid resistance has been documented in all the four subcounties (Ochomo et al., 2014). A map of the study sites is shown in figure 7. Figure 7: Map of Kenya showing the study sub-counties of Bondo, Nyando, Rachuonyo and Teso. 22

34 3.2 Mosquito sampling and rearing Mosquito larvae were collected in October 2015 to February 2016 in the four sub-counties in western Kenya. Dippers were used to collect Anopheles larvae from different habitats including ponds, pot holes, rice fields, drainage channels and mud paths. Collected specimens were kept in plastic larval containers which were labeled and transported to the laboratory at the Kenya medical research institute/ center for global health research (KEMRI/CGHR) for rearing under standard conditions (Temperature of C, photo-period of 12:12 hours (light: dark) and humidity of 50-70%). One of the sites from where the larvae were collected is shown in Figure 8. 23

35 Figure 8: Mosquito larval sampling in a pool of water in Kaliwa, Teso sub-county, western Kenya The larvae from the field were transferred unto large trays and debris removed from the water. The larval samples from the same sites were pooled together. Anopheles larvae were sorted and the remaining unwanted species discarded (Figure 9). The samples were placed in plastic containers for transportation to the KEMRI CGHR insectary. 24

36 Figure 9: Sorting mosquito larvae in Kamenya, Rachuonyo sub-county, before transportation to the KEMRI CGHR centre On arrival at the insectary at KEMRI CGHR the samples were sorted into different larval stages according to the instars. Similar instars were grouped together and placed in one tray. They were labeled in accordance to the site and collection date. The larvae were fed on powdered TetraMin fish food and brewer s yeast in a ratio of 3:1. Pupae were picked and placed in pupal cups inside the net cages. The emerged adult mosquitoes were fed on 10% sugar solution and the females exposed to different bioassays after two days. 25

37 3.3. Status of permethrin resistance in malaria vectors in western Kenya The status i.e. presence (resistant) or absence (susceptible) of permethrin resistance in Anopheles mosquitoes from the four sub-counties in western Kenya namely Bondo, Nyando, Rachuonyo and Teso was assessed using the WHO tube bioassay. Female Anopheles mosquitoes were exposed to permethrin (0.75%) pre-impregnated papers obtained from Universiti Sains, Malaysia. Approximately 100 female mosquitoes from each site were exposed to permethrin impregnated papers in four replicates. About female mosquitoes were aspirated into each exposure tube through the filling hole on the slide. These exposure tubes were set in a vertical position with the mesh screen end uppermost for a period of 1 hour after which the mosquitoes were transferred to a holding tube for 24 hours. A pad of cotton wool soaked in sugar solution was placed on the mesh screen end of the holding tubes. When Knock down Resistance (kdr) is involved, the rate of knock down is a quick indicator for early detection of resistance (WHO 2013b). Observation of the number of knocked down mosquitoes were made during the hour long exposure period. Mosquitoes were considered knocked down if they were unable to stand or fly in a coordinated way. The knocked down mosquitoes fell on the bottom of the exposure tube. Observations were made at an interval of 10 minutes for an hour before they were transferred to the holding tube. The mosquitoes were fed on 10% sugar solution until mortality was recorded after 24 hours. Counts of the number of dead mosquitoes were made after 24 hours post exposure. Mosquitoes were classified as dead if they were immobile or unable to stand or fly. If the mosquitoes lost their legs due to the excitorepelancy nature of permethrin some hours after exposure, but were still able to fly after the 24 hours, they were counted as being alive. 26

38 3.4. Intensity of Permethrin Resistance in malaria vectors in western Kenya The intensity of permethrin resistance in Anopheles mosquitoes from the four sub-counties of western Kenya namely Bondo, Nyando, Rachuonyo and Teso in western Kenya was evaluated using different permethrin concentrations with the CDC bottle bioassays Preparation of resistance diagnostic treatments A stock solution of four different concentrations ( 1, 2, 5 and 10) of permethrin insecticide was prepared. Four premeasured vials containing permethrin which when diluted in acetone and applied to the 250ml Wheaton bottles give insecticide concentrations of 1, 2, 5 and 10 (21.5µg/ml, 43µg/ml, 107.5µg/ml and 215µg/ml respectively) were provided by the CDC, Atlanta. To each insecticide vial, 500µl of acetone was added; this was then added to a 50ml falcon tube. A further 48ml of acetone was added to make the stock solution. The insecticides were labeled and stored at 4 C in the dark until use. The 250ml Wheaton bottles were marked using masking tape and numbered. The bottles were numbered, 1-4, for replication. The process was repeated for all the four concentrations (21.5µg/ml, 43µg/ml, 107.5µg/ml and 215µg/ml). The insecticide containing 21.5µg/ml was added to the first batch of four bottles, 43µg/ml was added to the second batch containing four bottles. This was done for 107.5µg/ml and 215µg/ml concentrations as well. The bottles were swirled to ensure that the entire inner surface and the lid were evenly covered with the insecticide. This was done until no trace of liquid was visible; they were then left to stand for an hour before the mosquitoes were introduced for exposure Exposing mosquitoes to resistance diagnostic treatments 27

39 Approximately female mosquitoes were aspirated and transferred into each bottle. The bottles were examined at time 0 where the number of dead and live was counted and recorded. Dead mosquitoes were recorded at an interval of 10 minutes for 2 hours after which the experiment was discontinued Effect of intensity of Permethrin Resistance on efficacy of insecticide treated nets Cone bioassays were used to test the effect of intensity of insecticide resistance on the efficacy of commonly used long lasting insecticidal nest (LLINs). The three different LLIN brands that were used for this bioassay included PermaNet 2.0, PermaNet 3.0 and Olyset. The LLINs were provided by the manufacturers: PermaNet 2.0 and PermaNet 3.0 were provided by Vestergaard Frandsen while Olyset net was provided by Sumitomo chemicals. According to the manufacturer the fabrics are impregnated with 55mg/m² deltamethrin for PermaNet 2.0 and 20g/kg permethrin for Olyset net. PermaNet 3.0 is impregnated with deltamethrin and an additional synergist, piperonylbutoxide is incorporated on the upper square. The roof contains 25g/kg of PBO in addition to 4g/kg of deltamethrin while the side panels are impregnated with 2.8g/kg deltamethrin (Yewhalaw et al., 2012) Performance of Cone Bioassays A piece of net measuring 25cm 25cm was cut from the four sides and the top of each net for replication. These nets were used in the cone bioassay to assess the mortality of the mosquitoes. A total of 20 female mosquitoes of age 2-5 days, non-blood fed, were exposed to the netting material where five mosquitoes were introduced in each cone for all the four cones. The mosquitoes were exposed to the net for 3 minutes after which they were 28

40 transferred to a paper cup where they were monitored and the number of knocked down individuals counted at an interval of 10 minutes for an hour. After the 1 hour the mosquitoes were fed on sugar solution and the overall mortality recorded after 24 hours Identification of Anopheles gambiae complex mosquitoes into species After performing the assays standard molecular assay was used to identify the field-collected siblings of An. gambiae s.l in to species (Collins et al., 1987, Scott et al., 1993) Extraction of mosquito genomic DNA Deoxyribonucleic acid (DNA) was extracted from whole mosquitoes using the alcohol precipitation method (Collins et al., 1987). The mosquitoes were stored individually in labeled vials in the freezer at -20 while waiting for the N extraction. Grinding buffer (100μl) was added to vials containing individual mosquitoes after which the mosquitoes were crushed to lyse and homogenize the cells. Grinding buffer consisted of homogenizing buffer and lysis buffer. Homogenizing buffer was made of 0.1 M sodium chloride, 0.2M sucrose, 0.01M ethylenediaminetetraacetic acid (EDTA) and 0.03M Tris Base. Lysis buffer consisted of 0.25M EDTA, 2.5% w/v sodium dodecyl sulfate (SDS) and 0.5M Tris Base. The resulting grinding buffer was the two combined in a ratio of 4:1 (Homogenisation: Lysis). The samples were then incubated in a 65 water bath for 30 minutes before adding 14μl of potassium acetate and mixing by vortexing. The mixture was incubated in ice cold water for 10 minutes then centrifuged at 13,200 rpm. A duplication of the labeled tube was done and after the revolution the supernatant was transferred to these new tubes and 200μl of ethanol added. These were stored at -20 for 20 minutes. The solution was then spun at 13,200 rpm for 20 minutes to allow the DNA to pellet. The vials were washed in 200μl 70% ethanol, then in the same volume of absolute ethanol. They were then inverted to dry overnight. The following morning 100μl of TE (Tris and EDTA) buffer was added to reconstitute the DNA. TE buffer consisted of 0.001M EDTA, 0.01M Tris-HCL at ph

41 Extracted DNA was then ready for use in subsequent testing of the species identity of the mosquitoes from which the DNA was obtained PCR amplification The normal polymerase chain reaction (PCR) was used to distinguish between An. gambiae and An. arabiensis (Scott et al., 1993). The reaction involved the use of two oligonucleotide primers that hybridized to opposite strands and flank the target DNA sequence that was to be amplified. The preparation of the PCR master mix and reaction mixture calculations was done as described below in table 1. Table 1: PCR master mixture Reagent Volume for 1 sample(µl) Volume for 100 samples(µl) Water PCR buffer MgCl₂ dntp mix Primer GA Primer AR Primer UN Taq DNA Polymerase Total The primers that were used included (a) a forward Universal primer (UN) (5 -GCT GCG AGT TGT AGA GAT GCG-3 ), (b) areverse An. gambiae primer (AG) (3 GCT TAC TGG TTT GGT CGG CAT GT-5 ), and (b) areverse An. arabiensis primer (AR) (3 -GTG TTA AGT GTC CTT CTC CGT C-5 ) ( N chemistry section, Biotechnology core facility branch, Division of scientific resources, Centres for Disease Control and Prevention, Atlanta, 30