The tsetse fly and its effects on agriculture in sub-saharan Africa

Tsetse fly — the enemy of people and animals of Africa

B.S. Hursey and J. Slingenbergh

The authors are Senior Officer (Insect-borne diseases) and Animal Health Officer (Trypanosomiasis), respectively, Animal Health Service, Animal Production and Health Division, FAO, Rome, Italy.

Tsetse flies, through the cyclical transmission of trypanosomiasis to both humans and their animals, greatly influence food production, natural-resource utilization and the pattern of human settlement throughout much of sub-Saharan Africa. It is estimated that the annual direct production losses in cattle alone amount to between US$6 000 million and $12 000 million, while animal deaths may reach 3 million.

The FAO Programme for the Control of African Animal Trypanosomiasis and Related Development was inaugurated during the World Food Conference, in 1974. Initially based on the concept of tsetse eradication from large tracts of sub-Saharan Africa, it has been subjected to significant revision and redirection. These changes have been influenced by two main factors: the development and introduction of new, more refined and environmentally acceptable control techniques (FAO, 1992, 1993) and an increasing awareness of the need to relate disease management to demography, population dynamics, natural-resource potential and agricultural systems. In this way, the basis necessary for strategic planning, at both the national and regional levels, is provided.

In order to carry out the required analysis, and to help clarify the tsetse problem at the subcontinental level, FAO has initiated the development of a geographic information system (GIS) on tsetse and agriculture.

The underlying principle is, in itself, quite simple. There are 37 tsetse-infested countries in sub-Saharan Africa. The disease manifests itself when and where humans and their livestock are placed at risk of infection. Indirect problems also result from efforts to avoid tsetse contact, however. These may be readily demonstrated by the subcontinental cattle distribution. Of the 165 million cattle in sub-Saharan Africa, only 10 million are located in tsetse-infested areas, while the remainder are distributed on the periphery. The difficulty lies in the interpretation of the consequences of this distorted distribution, which, in turn, depends on a knowledge of the factors determining cattle distribution patterns in the absence of tsetse.

In this regard, significant progress has been made by the Environmental Research Group, Oxford (ERGO), who examined the anthropogenic and environmental correlates of livestock distribution in the West African semi-arid and subhumid ecological zones (Wins and Bourn, 1994). This study demonstrated that human habitation patterns and the associated intensity of crop production, as well as rainfall, are all key factors influencing the distribution of ruminant livestock. In general, there is a tendency towards the aggregation of people, crops and livestock. This holds true for both pastoral and village cattle when they are kept on a year-round basis. Cattle densities are highest in the moister subhumid areas, except where tsetse flies exist. This conclusion is of major relevance to the understanding of pastoral and agropastoral production systems as it contradicts the conventional belief that livestock are generally kept in drylands, away from the moister cropping zones.

Screw worm at what price?

The eradication of New World Screwworm-induced myiasis from the Libyan Arab Jamahiriya in 1992, at a cost of US$75 million, yielded an estimated cost-benefit ratio of 50:1 when calculated against the value of the livestock and wildlife in the region overall. Similarly, the ongoing 30-year campaign to eradicate Screw worm from the United States and Mexico, which is to eventually extend south to the Darien gap, so far has cost some $700 million, but it has given an average cost-benefit return to the livestock industry of 6:1.

In Australia, it has been estimated that the invasion of the Old World Screw worm could incur annual livestock losses of up to A$430 million.

Various myiasis-causing insects inhabit vast areas throughout the world, and at what cost to the subsistence farmers and rural poor who struggle to survive in the face of these devastating parasites? Following the success of the Libyan campaign, FAO intends to undertake a global assessment of the impact of myiases in order to set priorities for intervention and to better advise Member Nations.

Recognition of the fact that the ultimate goal of sustainable rural development depends on healthy and productive livestock is the driving force behind this initiative. And the availability of these livestock demands that major constraints, such as the Screw worm be identified and systematically addressed.

The ERGO data were kindly made available to FAO for more extensive analysis in the context of defining the impact of tsetse on livestock and agriculture. These studies revealed that, in the tsetse-infested areas of moist subhumid regions of Nigeria, village cattle are virtually absent, despite high population density, land pressure, intensive cereal cropping and a favourable length of growing period. Therefore, in otherwise identical agro-ecological sets of circumstances, it is the presence of tsetse alone that prohibits the keeping of cattle on a more permanent basis. This was corroborated by relating cattle densities to the amount of woody vegetation as measured by satellite imagery. The results confirmed that the discrepancy between livestock densities in tsetse-infested and non-infested areas is explained by the presence of the fly, as well as by the disease risk it poses to humans and animals, rather than by the mere fact that woody vegetation tends to coincide with tsetse distribution (Rogers, Hendrickx and Slingenbergh, 1994).

A computer simulation model was designed to forecast the patterns of human and cattle population densities and those of arable land use in the moist subhumid zone of Nigeria. The data have since been transferred to GIS for mapping purposes. Some preliminary outputs of this model are shown in Figures 1 to 4.

The model predicts the results of interactions between such variables as human and cattle populations, their growth rates and the intensity of cultivation. It is also able to take into account the influence of tsetse/trypanosomiasis on these interactions. At this stage, the environmental impacts of arable land expansion and increased human and livestock populations, for example, a reduction in woody vegetation, have not yet been included. Their importance is recognized, however, and basic interrelationships between the main variables affecting the environment can be readily incorporated into the model by using baseline normalized differential vegetation index (NDVI) data obtained through satellite imagery (Hendrickx et al ., 1995).

More important, the model depicts the areas where agricultural expansion is the most significant and the need for ruminant livestock integration is the greatest. It is the persistence of tsetse, particularly the riverine species, around these areas that causes major damage. Fortunately, the technical capacity to suppress the vector as well as the incentives to activate the local communities, in terms of more rewarding farming practices, are greatest in this particular set of circumstances.

A further indication of the significant influence of trypanosomiasis on livestock distributions has been obtained from an FAO field project in Togo. In recent years, this project has systematically collected all relevant data required for a trypanosomiasis impact analysis. The findings shed new light on the influence of tsetse on farming systems. As for Nigeria, while the number of rural household cattle increases proportionally with the amount of land brought into the cultivation cycle, the positive association between cropping and cattle becomes weaker as exposure to trypanosomiasis increases. In areas with a trypanosomiasis prevalence of more than 30 percent, it becomes virtually impossible to establish and maintain a mixed-farming system.

The influence of trypanosomiasis risk on the distribution of various cattle breeds was also examined by the Togo project, which reaffirmed that the trypanotolerant Baoulé are significantly more resistant to the disease than zebu. It was also established that the degree of disease resistance, measured on a herd basis, strongly correlated with the number of Baoulé genes as estimated from the phenotypes. Moreover, the study suggested that zebu introgression rapidly reduced the breed’s capacity to survive under tsetse challenge. It is noteworthy that livestock owners have appreciated this, and, as a result, the distribution of Togolese cattle breeds is very much a function of the level of disease risk to which they are exposed: 50 percent of the herds consist entirely of purebred Baoulé animals. The implication is that the integration of crop and livestock production develops very gradually because of the risk involved in introducing zebu animals, while the Baoulé cattle lack the size and strength required to provide adequate draught power.

A comprehensive epidemiological study of this nature facilitates the identification of effective strategies for disease intervention. In the Togo example, such strategies were established in accordance with the diverse disease-risk situations and the degree of innate resistance found in the herds. In areas where animals are mostly tolerant and the disease risk is low, the problem may be contained by treating only those individual animals affected with drugs. Whereas, at the other extreme, where trypanosusceptible zebu are exposed to high risk, the intervention of choice would be risk reduction through vector control. The intermediate scenarios, which more closely approximate the practical situation, would demand the appropriate mix of control methods.

In West Africa, this constraint to the development of mixed farming is most severe in the areas recently freed from onchocerciasis. The significant reoccupation of fertile valleys has triggered a major expansion of agricultural production. It is also evident that the presence of tsetse prohibits effective and productive use of natural resources, however, by constraining the integration of livestock with crop production. The alleviation of this constraint is a matter 1of some urgency because of the extremely high and increasing demand for land resulting from the demographic explosion, a situation that is particularly acute in this subregion. These high population pressures also influence the movement of traditional pastoral societies away from the Sudano-Sahelian zone. For some decades, this increasing pressure has been forcing pastoralists to move from the growing intensity of cultivation in the moister regions of the arid and semi-arid zones towards the underutilized tsetse-infested areas. However, as the majority of pastoralist cattle are trypanosusceptible zebu, which cannot produce or survive under the constant risk of trypanosomiasis, the results are often disastrous.

An analysis made to identify priority areas for intervention indicates that, in the medium term, more sustainable socio-economic benefits are to be gained by focusing on the wetter mixed-farming areas of the subhumid zone, especially where human populations are at their greatest. A major consideration when implementing vector control in such situations is the readiness of rural societies to play an active role in sustaining the tsetse campaign long enough to achieve autonomous disease control through the gradual and progressive transformation of the landscape.

Tsetse-transmitted trypanosomiasis is unique in that it has considerable impact over a vast area of some 9 million km 2 , where it so profoundly influences and distorts the patterns and density of agriculture that they are often contrary to the demand and the resource potential. Realizing the magnitude of this problem, the programme is in the process of changing towards a more realistic and practical approach. Most notably, this includes strengthening the coordinating role of FAO within a global programme to clarify and solve the problem of African trypanosomiasis, both human and animal, and to ensure the active participation of all players, from the subsistence farmer to the international research institutes and donor community (FAO, 1995; FAO/IBAR, 1995). It is now generally appreciated that strategic planning for tsetse and trypanosomiasis cannot be undertaken within the narrow framework of animal health, nor, indeed, confined to the livestock sector alone. It requires an understanding of resource potential, environmental implications, farming systems and the constraints thereon, as well as consideration of the dynamics of population growth and food demand over time. It has rural development dimensions.

See also:  Should You Use Liquid Soap to Remove Ticks?

FAO . 1992. Training manual for tsetse control personnel Vol. 4: Use of attractive devices for tsetse survey and control . Rome, FAO. 196 pp.

FAO . 1993. Training manual for tsetse control personnel . Vol. 5: Insecticides for tsetse and trypanosomiasis control using attractive bait techniques . Rome, FAO. 88 pp.

FAO . 1995. The coordination of research and development within a global programme to clarify and solve the trypanosomiasis problem . Vienna, Austria, Joint FAD/IAEA Division.

FAO & Interafrican Bureau for Animal Resources (IBAR). 1995. The global programme to clarify and solve the trypanosomiasis problem . Inaugural meeting of the Steering Committee. Rome, FAO.

Everything You Need to Know About African Sleeping Sickness

TripSavvy / Nez Riaz

Many of Africa’s most notorious diseases are transmitted by mosquitoes — including malaria, yellow fever and West Nile virus. However, mosquitoes aren’t the only potentially deadly insect on the African continent. Tsetse flies transmit African trypanosomiasis (commonly known as sleeping sickness) to animals and humans in 36 sub-Saharan countries. Infection is usually confined to rural areas and is therefore most likely to affect those planning on visiting farms or game reserves.

The Tsetse Fly

The word «tsetse» means «fly» in Tswana, and refers to all 23 species of the fly genus Glossina. Tsetse flies feed on the blood of vertebrate animals (including humans) and in doing so, transmit the sleeping sickness parasite from infected animals to uninfected ones. The flies resemble normal house flies, but can be identified by two distinguishing characteristics. All tsetse fly species have a long probe, or proboscis, extending horizontally from the base of their head. When resting, their wings fold over the abdomen, one exactly on top of the other.

Sleeping Sickness in Humans

Of the 23 tsetse fly species, only six transmit sleeping sickness to people. There are two strains of human African trypanosomiasis: Trypanosoma brucei gambiense and Trypanosoma brucei rhodesiense. The former is by far the most prevalent, accounting for 97% of reported cases. It is confined to Central and West Africa and can go undetected for months before serious symptoms emerge. The latter strain is less common, faster to develop and confined to Southern and East Africa. Uganda is the only country with both T.b. gambiense and T.b. rhodesiense.

Symptoms of sleeping sickness include fatigue, headaches, muscle aches and a high fever. In time, the disease affects the central nervous system, resulting in sleep disorders, psychiatric disorders, seizures, coma and eventually, death. Fortunately, sleeping sickness in humans is on the decline. According to the World Health Organization, the number of reported cases dropped below 10,000 for the first time in 50 years in 2009. In 2015, only 2,804 new cases were reported. The decline is attributed to better control of tsetse fly populations, as well as improved diagnosis and treatment.

Avoiding Sleeping Sickness

There are no vaccines or prophylactics for human sleeping sickness. The only way to avoid infection is to avoid getting bitten — however, if you are bitten, the chances of infection are still small (less than 0.1%). If you plan on traveling to a tsetse-infected area, make sure to pack long-sleeved shirts and long pants. Medium-weight fabric is best because the flies can bite through thin material. Neutral tones are essential as the flies are attracted to bright, dark and metallic colors (and especially blue — there’s a reason that safari guides always wear khaki).

Tsetse flies are also attracted to moving vehicles, so make sure to check your car or truck before starting a game drive. They shelter in dense bush during the hottest hours of the day, so schedule walking safaris for the early mornings and late afternoons. Insect repellent is only marginally effective in warding the flies off. However, it’s worth investing in permethrin-treated clothing and repellent with active ingredients including DEET, Picaridin or OLE. Make sure that your lodge or hotel has a mosquito net or pack a portable one in your bag.

Treating Sleeping Sickness

Keep an eye out for the symptoms listed above, even if they occur several months after you return from a tsetse-infected area. If you suspect that you may have been infected seek medical attention immediately, making sure to tell your doctor that you have recently spent time in a tsetse country. The drugs that you will be given depend on the strain of tsetse you have, but in either case, it is likely that you will need to be screened for up to two years to ensure that the treatment has been successful.

Likelihood of Contracting Sleeping Sickness

Despite the severity of the disease, you shouldn’t let the fear of contracting sleeping sickness stop you from coming to Africa. The reality is that tourists are unlikely to get infected as those most at risk are rural farmers, hunters and fishermen with long-term exposure to tsetse areas. If you’re worried, avoid traveling to the Democratic Republic of the Congo (DRC). 70% of cases originate from here and it is the only country with more than 1,000 new cases annually.

Popular tourist destinations like Malawi, Uganda, Tanzania and Zimbabwe all report fewer than 100 new cases each year. Botswana, Kenya, Mozambique, Namibia and Rwanda haven’t reported any new cases in over a decade, while South Africa is considered sleeping sickness-free. In fact, South Africa is a good choice for anyone worried about insect-borne diseases as it has a wide choice of game reserves that are also malaria-free.

Sleeping Sickness in Animals

Animal trypanosomiasis has a devastating effect on livestock, especially cattle. Infected animals become increasingly weak and cannot plough or produce milk. Pregnant females often abort their young and eventually, the victim will die. Prophylactics for cattle are expensive and not always effective. As such, large-scale farming is impossible in tsetse-infected areas. Those that do attempt to keep cattle are plagued by sickness and death, with approximately 3 million cattle dying every year from the disease.

Because of this, the tsetse fly is one of the most influential creatures on the African continent. It is present in an area spanning approximately 10 million square kilometers of sub-Saharan Africa — fertile land that cannot be successfully farmed. As such, the tsetse fly is often labeled as one of the major causes of poverty in Africa. Of the 36 countries affected by animal African trypanosomiasis, 30 are ranked as low-income, food-deficit nations.

On the other hand, the tsetse fly is also responsible for preserving vast tracts of wild habitat that would otherwise have been converted to farmland. These areas are the last strongholds of Africa’s indigenous wildlife. Although safari animals (especially antelope and warthog) are vulnerable to the disease, they are less susceptible than cattle.

Seasonal variation of tsetse fly species abundance and prevalence of trypanosomes in the Maasai Steppe, Tanzania

School of Life Sciences and Bioengineering, The Nelson Mandela African Institution of Science and Technology, Arusha 477, Tanzania

Department of Geography and Environmental Studies, University of Dodoma, Dodoma, Tanzania

School of Life Sciences and Bioengineering, The Nelson Mandela African Institution of Science and Technology, Arusha 477, Tanzania

Department Conservation Biology, University of Dodoma, Dodoma, Tanzania

Genome Science Centre and Department of Microbiology, Parasitology and Immunology, Sokoine University of Agriculture, Morogoro, Tanzania

School of Life Sciences and Bioengineering, The Nelson Mandela African Institution of Science and Technology, Arusha 477, Tanzania

Center for Infectious Disease Dynamics, Huck Institutes of the Life Sciences and Department of Biology, Pennsylvania State University, University Park, PA 16802 U.S.A.

Center for Infectious Disease Dynamics, Huck Institutes of the Life Sciences and Department of Biology, Pennsylvania State University, University Park, PA 16802 U.S.A.

School of Life Sciences and Bioengineering, The Nelson Mandela African Institution of Science and Technology, Arusha 477, Tanzania

Genome Science Centre and Department of Microbiology, Parasitology and Immunology, Sokoine University of Agriculture, Morogoro, Tanzania

Center for Infectious Disease Dynamics, Huck Institutes of the Life Sciences and Department of Biology, Pennsylvania State University, University Park, PA 16802 U.S.A.

School of Life Sciences and Bioengineering, The Nelson Mandela African Institution of Science and Technology, Arusha 477, Tanzania

Department of Geography and Environmental Studies, University of Dodoma, Dodoma, Tanzania

School of Life Sciences and Bioengineering, The Nelson Mandela African Institution of Science and Technology, Arusha 477, Tanzania

Department Conservation Biology, University of Dodoma, Dodoma, Tanzania

Genome Science Centre and Department of Microbiology, Parasitology and Immunology, Sokoine University of Agriculture, Morogoro, Tanzania

School of Life Sciences and Bioengineering, The Nelson Mandela African Institution of Science and Technology, Arusha 477, Tanzania

Center for Infectious Disease Dynamics, Huck Institutes of the Life Sciences and Department of Biology, Pennsylvania State University, University Park, PA 16802 U.S.A.

Center for Infectious Disease Dynamics, Huck Institutes of the Life Sciences and Department of Biology, Pennsylvania State University, University Park, PA 16802 U.S.A.

School of Life Sciences and Bioengineering, The Nelson Mandela African Institution of Science and Technology, Arusha 477, Tanzania

Genome Science Centre and Department of Microbiology, Parasitology and Immunology, Sokoine University of Agriculture, Morogoro, Tanzania

Center for Infectious Disease Dynamics, Huck Institutes of the Life Sciences and Department of Biology, Pennsylvania State University, University Park, PA 16802 U.S.A.


Tsetse flies, the vectors of trypanosomiasis, represent a threat to public health and economy in sub‐Saharan Africa. Despite these concerns, information on temporal and spatial dynamics of tsetse and trypanosomes remain limited and may be a reason that control strategies are less effective. The current study assessed the temporal variation of the relative abundance of tsetse fly species and trypanosome prevalence in relation to climate in the Maasai Steppe of Tanzania in 2014–2015. Tsetse flies were captured using odor‐baited Epsilon traps deployed in ten sites selected through random subsampling of the major vegetation types in the area. Fly species were identified morphologically and trypanosome species classified using PCR. The climate dataset was acquired from the African Flood and Drought Monitor repository. Three species of tsetse flies were identified: G. swynnertoni (70.8%), G. m. morsitans (23.4%), and G.pallidipes (5.8%). All species showed monthly changes in abundance with most of the flies collected in July. The relative abundance of G. m. morsitans and G. swynnertoni was negatively correlated with maximum and minimum temperature, respectively. Three trypanosome species were recorded: T. vivax (82.1%), T. brucei (8.93%), and T. congolense (3.57%). The peak of trypanosome infections in the flies was found in October and was three months after the tsetse abundance peak; prevalence was negatively correlated with tsetse abundance. A strong positive relationship was found between trypanosome prevalence and temperature. In conclusion, we find that trypanosome prevalence is dependent on fly availability, and temperature drives both tsetse fly relative abundance and trypanosome prevalence.


The distribution and abundance of vectors is determined by the interplay of three factors: suitable climatic conditions, habitat for development, and the availability of hosts for food. These factors are not independent, since changes in climate not only directly affect the conditions for vector development but also indirectly alter vegetation cover and the movement of the hosts (Jones et al. 2007 , Reisen et al. 2008 , Mills et al. 2010 , Moore and Messina 2010 , Srimath‐Tirumula‐Peddinti et al. 2015 ). Of the different aspects of climate, temperature has been shown to influence the growth and proliferation of trypanosomes within the tsetse fly (Walshe et al. 2009 ). There are multiple approaches that can be used to examine the effects of climate variation on vector distribution and parasite developments (Moore et al. 2011 ). In this paper, we examine the within‐year fluctuations in temperature and rainfall in the Maasai Steppe of Tanzania and how these are associated with changes in relative abundance of three Glossina species and the prevalence of trypanosomes within them, which in turn affects the likelihood of cattle and humans becoming infected with trypanosomiasis.

See also:  The World s Most Dangerous Ants

The general consensus among infectious disease ecologists is that changes in climate alter the distribution of many infectious diseases (Gray et al. 2009 , Moore et al. 2011 , Huynen and Martens 2013). Climate envelope models have been used to examine the distribution of vectors, as well as calculation of vector vital rates, including transmission rates (Epstein 2001 , Anderson et al. 2004 , Rödder et al. 2008 ). Correlations between climatic variables and abundance have also been used to examine the impact of changes in climate on vector‐borne diseases (Lafferty 2009 , Moore et al. 2011 , Paaijmans et al. 2012 , Mordecai et al. 2013 ). Both average climatic conditions and day‐to‐day variation in temperature are known to be important for vector and parasite development (Hargrove 2004 , Patz et al. 2005 , Kleynhans and Terblanche 2011 , Lukaw et al. 2014 ). For these reasons, studies involved in monitoring abundance and prevalence of vectors and pathogens, coupled with records of location and climate, have been used to provide insights that can assist in describing the relationship between climate variation and vector and pathogens dynamics, and thus an inference of seasons and areas at risk of disease.

Tsetse flies have been recorded in more than three quarters of the Tanzania rangelands with high abundance in wildlife protected areas and adjacent zones (Lucas et al. 2001 , Malele et al. 2011 ). These flies are vectors of both human and animal African trypanosomiasis, and play a significant role in compromising the health of people and livestock and economic development (Malele 2011 ). Currently, trypanosomiasis threatens the livestock sector in Tanzania, accounting for about 4.7% of the National GDP and 13% of the Agricultural GDP (Tumbo et al. 2011 ). The infection also threatens more than four million rural Tanzanians, including the Maasai communities (Malele et al. 2006 , Malele et al. 2011 ). Maasai communities are particularly vulnerable to negative impacts of trypanosomiasis since their economic status and nutrition are highly tied to livestock production. These communities are also challenged by current land tenure, climate change, and a number of other livestock infections. In addition, information on trypanosomiasis risk in Maasai areas is limited, and this compromises vector control programs.

Recent studies on tsetse fly and trypanosomiasis risk have mainly focused on biological aspects of the vector and the protozoa within the Maasai Steppe, and there remains a lack of knowledge about climate and its influence on risk of infection (Malele et al. 2006 , Matemba et al. 2010 , Salekwa et al. 2014 , Muse et al. 2015 ). Maasai Steppe consists of continuous savanna grassland dotted with trees and is home to the Maasai people, who are traditional pastoralists. This study was carried out in the Masaai Steppe of Tanzania and addressed the following questions: (1) Do tsetse fly abundance and trypanosome prevalence vary between months? (2) Is seasonal variation of climate associated with tsetse fly abundance and pathogen prevalence? (3) Does tsetse fly abundance influence trypanosome prevalence?


The tsetse fly sampling was undertaken in Emboreet village in the Simanjiro District, part of the Tanzanian Maasai Steppe located between 4° 47’ 15” S and 36° 53’ 54” E (Figure 1). The area is semi‐arid, characterized by grassland savanna with Acacia woodlands, Commiphora species, scattered baobab (Adansonia digitata), and sausage (Kigelia africana) trees. The area is inhabited by the Maasai people who herd livestock and co‐exist and share pasture with free‐ranging wild ungulate species. Surface water is scarce and seasonal availability determines the movement and presence of livestock and the wild animals. Economic activities by the Maasai focus primarily on livestock and more recently agricultural production (McCabe et al. 2010 ). Eco‐tourism is important in this part of Tanzania and there are several national parks (NP), including Tarangire and Lake Manyara NP.

Tsetse fly abundance, trypanosome prevalence, and climate data collection

Temporal abundance of tsetse flies was estimated by monthly sampling for a period of 15 months in 2014 and 2015 in the village of Emboreet, which borders Tarangire National Park (Figure 1). Sites were selected through stratified random subsampling of the major vegetation types in the area (Bouyer et al. 2013 ). A total of ten sites were identified and three epsilon traps were deployed at each site and located at least 200 m apart (Malele et al. 2011 ). At each trap, the grass vegetation was cut to ground level and the legs of the trap greased to avoid ants consuming caught flies. Each trap was baited with an attractant made from acetone, phenols, and octanol (Hargrove et al. 1995 ). Traps were checked and emptied every 24 h for six days each month, at which time the number of individuals for each tsetse species within each trap was recorded, along with the GPS (global positioning system) location. Tsetse flies were identified using training manuals published by the Food and Agricultural Organization (Pollock 1982 ) and identification was confirmed with a qualified entomologist from the Vector and Vector‐Borne Diseases Research Institute located in Tanga, Tanzania. After six days, the traps were removed from the site and deployed again for sampling the following month.

Collected flies were preserved individually in Eppendorf tubes filled with absolute ethanol. DNA was extracted from individual flies in the laboratory using an ammonium acetate precipitation protocol (Bruford et al. 1998 ). All DNA samples were stored at −20° C until further analysis. Trypanosome species were identified by polymerase chain reaction (PCR) undertaken on fly‐specific pools (Ferreira et al. 2008 , Malele et al. 2013 ). Each pool was prepared by mixing ten individual DNA samples in equal volumes. Positive pools of DNA samples were further studied to identify the individual flies that were infected in order to establish prevalence of trypanosome species within each tsetse species. A total volume of 15 μl containing 7.5 μl Dream Taq master mix, 200 nM of forward and reverse primers, and 3.9 μl of nuclease‐free water was used during the PCR amplification process where the ITS1 gene was targeted. The PCR products were separated on 2% GR green‐stained agarose gels and positive results were identified based on PCR product size corresponding to 300 bp for T. vivax, 400 bp for T. brucei, and 700 bp for T. congolense savannah. T. brucei‐positive samples were further tested using SRA gene amplification primers in order to identify human infective trypanosomes. Detailed primer sequences and cycling conditions followed in this study are described in Radwanska et al. ( 2002 ) and Njiru et al. ( 2005 ).

Climate data were acquired from the African Flood and Drought Monitor repository which provides satellite gridded data at spatial and temporal resolution of 0.25° at daily intervals, ( Daily, maximum and minimum temperature, and total precipitation for the year 2014 and 2015 were acquired and used to characterize the weather of the study area. After acquisition, data were checked for quality before being used in the analysis. These data have limitations of spatial resolution since the trapping sites fall into only two grids of satellite climatic data. Availability of local station data on‐the‐ground or alternative high resolution data for the study area is poor and/or limited.

Data analysis

The relationships between climate and tsetse fly abundance or trypanosome prevalence, and also between tsetse flies and trypanosomes, were examined using linear mixed effect models (LME) with the ‘lme4’ package in the statistical program R (v. 3.24) (Baayen et al. 2008 ). When possible, the contribution of independent variables with quadratic or exponential functions, as well as two‐way interactions with other independent variables, was also examined. Sampling site was included as a random factor to take into account between‐site variability and the re‐sampling of the same site over the months. The minimum parsimonious model was considered and presented in our results. To examine the effect of time lag in relative abundance of tsetse fly species on prevalence, cross correlation analysis was implemented and a delay of up to three months was examined using the ‘astsa’ package in R (Rao 2001 ).


While our goal was to examine the seasonality in the relative abundance of tsetse flies and their infection rate, a preliminary analysis was performed to investigate the role of habitat in the temporal variation of fly catches. Analysis confirmed that habitat significantly affected changes in total fly abundance at the monthly level. Given that the resolution of our climatic data covered the habitat selected in our study sites, our analyses were performed combining the habitat together and using the sampling site as a random factor in the analysis to take into account spatial variability in fly trapping and trypanosome prevalence.

During the study period, the lowest temperature was recorded in July and the highest in February. Rainfall was relatively variable between months, with the maximum amount of rainfall recorded in April and the minimum in September (Figure 2).

Temporal dynamics of tsetse fly relative abundance

A total of 3,000 tsetse flies, comprising three savanna Glossina species, were caught during the 15 months of trapping. G. swynnertoni (70.8%) was the most abundant species followed by G. m. morsitans (23.4%) and, at a much lower value, G. pallidipes (5.8%). Temporal changes in abundance were recorded in all three species. G. swynnertoni abundance peaked in July, followed by September and then March, while G. m. morsitans and G. pallidipes exhibited two peaks each: G. m. morsitans peaked in July and March, whereas G. pallidipes peaked in November and July (Figure 3). Differences in abundance of tsetse species were significant for a number of months (Table 1).

Climate‐tsetse fly relationships

A negative relationship was found between relative abundance of G. m. morsitans and maximum temperature and between G. swynnertoni and the minimum temperature, while relative abundance of G. pallidipes showed no associations with any of the temperature variables (Table 2).

MONTH G. pallidipes G. m. morsitans G. swynnertoni
Coef ± SE Df P Coef ± SE Df P Coef ± SE Df P
*Intercept 0.004 ±.02 2236 0.84 0.00001±.07 2236 1 0.04 ±.14 2236 0.79
February 0.009 ±.02 2236 0.62 0.007±.35 2236 0.8 0.21 ±.05 2236
G. m. morsitans G. swynnertoni
Coef± SE Df P Coef± SE Df P
*Intercept 21.2 ± 3.6 481 0.0000 **Intercept 6.32 ± 8.2 481 0.4
Tmax −7 ± 2.0 481 0.0004 Tmax 7.9 ± 4.5 481 0.08
Tmin 1.2 ± 2.0 481 0.5554 Tmin −10.8 ± 4.5 481 0.02
Random effect: Site 0.79 Random effect: Site 2.5
AIC value 2266.17 AIC value 3075.9

Temporal dynamics of trypanosomes prevalence

Most of the tsetse fly infections were from T. vivax (82.1%), with proportionately fewer from T. brucei (8.93%) and T. congolense (3.57%). Co‐infections with T.vivax and T. brucei were the most common (3.57%), while T. vivax‐T. congolense and T. vivax‐T. brucei‐T. congolense were rare and only 0.89% each. Further analysis of T. brucei positive flies found no human infective species, specifically T. brucei rhodensiense. The highest upper and interquartile values of trypanosome prevalence were recorded in October, while lowest quartiles and interquartiles of prevalence were recorded in January. These results suggest that October consistently scored the highest prevalence and January the lowest compared to the other months (Figure 4). In general, prevalence of trypanosomes increased from January to November with some monthly fluctuations, but only the peak in October was significant (Table 3).

MONTH Prevalence
Coef ± SE Df P
*Intercept 0.00 ± 4.8 63 1
February 0.83 ± 5.6 63 0.9
March 1.44 ± 5.6 63 0.8
April 1.25 ± 5.7 63 0.8
May 9.22 ± 5.6 63 0.1
June 9.26 ± 5.9 63 0.1
July 2.73 ± 5.2 63 0.6
August 1.49 ± 5.2 63 0.8
September 3.52 ± 5.2 63 0.5
October 12.20 ± 5.2 63 0.05
November 8.68 ± 5.7 63 0.1
Random Factor: Habitat 0.0003
AIC value 554.9

Relationship between climate and prevalence of trypanosomes

Following previous work, our initial prediction was that trypanosome prevalence would rise with temperature, reach a peak, and then decline (Cross and Manning 1973 ). Accordingly, we found that trypanosome prevalence increased with rising maximum temperatures from 26° C to 31° C, and declined beyond a maximum temperature of 31° C (Figure 5). This trend was described by a significant positive relationship of prevalence with maximum temperature linear (β = 23.14, SE= 9.07, df = 37, p=0.0150) and quadratic terms (β = −0.37, SE= 0.15, df = 37, p=0.0177).

Relationship between prevalence of trypanosomes and relative abundance of tsetse flies

The peak in trypanosome prevalence lagged three months behind the July peak in tsetse fly abundance (Figures 3 and 4) but this lag effect was not statistically significant. A negative relationship was found between trypanosome prevalence and tsetse fly abundance, suggesting that high prevalence is related to low fly abundance (Figure 6). In general, it was observed that an increase in one unit of tsetse fly abundance reduces prevalence of trypanosomes (β = −7.66, SE= 1.23, df = 38, p=0.001).


The relationship between seasonal variation of climate and tsetse fly abundance or trypanosome prevalence in the Maasai Steppe of Tanzania has not received adequate attention. Yet this area has abundant wildlife reservoir hosts of zoonotic diseases. It is also used heavily by livestock and can be a potential hotspot for infectious diseases of human concern, including the neglected tropical disease trypanosomiasis. In order to fill this gap in knowledge, the current study addressed this concern by performing a longitudinal sampling at Emboreet Village in 2014–2015.

This study identified G. swynnertoni and T. vivax as the most dominant tsetse fly and trypanosome species, respectively, during the sampling period. The study has clearly shown temporal variation in relative abundance of different species of tsetse flies that appeared to be mainly associated with a negative effect of minimum and maximum temperature. Furthermore, we have shown temporal variations of trypanosome prevalence that were significantly affected by maximum temperature. Trypanosome prevalence peaked three months after the peak in tsetse fly relative abundance, but a negative exponential relationship was found between tsetse fly relative abundance and trypanosome prevalence, where lower tsetse abundance is associated with a higher prevalence of trypanosomes.

The relatively large number of G. swynnertoni reported in this study confirms the findings by Sindato et al. ( 2007 ) and Salekwa et al. ( 2014 ) who also reported dominance of G. swynnertoni in the same area. Dominance of this species may suggest that even though the other two species, G. pallidipes and G. m. morsitans, can pose potential risk as vectors for trypanosomiasis, G. swynnertoni could still be the most important species in driving the epidemiology of infection in the area. Predominance of T. vivax was expected since it is known to be a widely distributed species due to both mechanical and cyclical transmissions (Dagnachew and Bezie 2015 ). Predominance of T. vivax around the Maasai Steppe has also been reported in other studies (Adams et al. 2010 , Swai and Kaaya 2012 ). Okoh et al. ( 2012 ) also found predominance of T. vivax from an investigation carried out in one of the national parks of Nigeria. Since persistence of T. vivax requires only one fly and three to four animal hosts (Rogers 1988 ), the presence of numerous wild ungulates and cattle in the study area further support the dominance of T. vivax. Further analysis of T. brucei‐positive samples did not indicate the presence of human infective trypanosomes, but these negative results should not support the notion that the Masaai Steppe is a sleeping sickness‐free zone. The lack of positive cases may be attributed to the fact that human infective trypanosomes are usually at low prevalence and often not captured in relatively low sample size. Auty et al. ( 2012 ) reported similar findings in a study carried out in the Serengeti NP.

The relative abundance of tsetse flies was greatest in July, whereas relatively low catches were recorded in January. In general, the months with high catches corresponded to a dry period. At this time of year, adult tsetse flies rely entirely on blood from available hosts (Hargrove 2004 ) and they appear to be well adapted to this dry environment. It is possible that the concentration of available animals around resources like water and food offers good opportunities to overcome the extreme dry climate while allowing the reproductive life cycle of the fly to continue. Indeed, in the dry months of the Maasai Steppe, wild animals and livestock congregate and feed on bushes where food and protection are available and tsetse flies have access to the hosts for blood. Since there is a frequent interaction between vectors and hosts, this period can pose a high risk for animal trypanosomiasis. The presence of relatively high abundance of flies during the dry seasons has also been reported by Lukaw et al. ( 2014 ) and Sindato et al. ( 2007 ). However, it is also possible that flies are more mobile during dry months and, thus, while the catch increases, the actual abundance might remain unchanged. Similarly, the availability of hosts may also increase during this period, as a consequence of resource driven animal movements, and when the hosts move through the trapping areas, the catch rate increases. For these reasons, the recorded variation in abundance does not necessarily reflect the abundance in the entire area but it gives an understanding of what to expect in different seasons within this area.

We found no strong relationship between climate parameters and G. pallidipes, but a significant negative relationship between minimum temperature and G. swynnertoni, and maximum temperature with G. m. morsitans was observed. The weak relationship between G. pallidipes and climate parameters could be associated with the ability of this species to thrive at low abundance in different areas and under mild climatic conditions (Pollock 1982 ). Also, the small sample size of G. pallidipes could have influenced these results. The significant negative relationship of G. swynnertoni and G. m. morsitans relative abundance with minimum and maximum temperature could be partly because both low and high temperatures reduce the activity of the flies. At cold temperatures, fly activity falls and then again at high temperatures as the flies seek refuges and are more likely to enter the traps (Torr and Hargrove 1999 , Terblanche et al. 2008 ). The observed disparities among the tsetse fly species and the climatic variables raises questions as to whether G. swynnertoni tolerance is greater to high temperatures and G. m. morsitans tolerance is greater to lower temperatures, however we do not have data or published work to test this hypothesis. Overall, these findings were counter to our expectation of temperature being a robust predictor of tsetse fly species abundance (Rogers et al. 1996 , Moore et al. 2011 ). Part of the weak signal observed could be associated with the low spatial resolution of the satellite climate data that failed to provide microclimate changes and variation during the trapping sessions.

The presence of trypanosomes throughout the year could be associated with three factors: first, the year‐round presence of wild animals as “trypanosome reservoirs” in areas close to the park; second, the evidence that adult tsetse flies feed entirely on blood, which allows the chance of year‐round trypanosome circulation in the area; third, adult tsetse flies appear to thrive well through much of the variation in weather conditions and thus there is a risk of transmission throughout the year (Hargrove 2004 ). The high trypanosome prevalence recorded in October was contrary to our expectation and the findings of previous studies in the same area (Sindato et al. 2007 ). Possible reasons for this observation could be the fact that October coincides with a period of short rains (Figure 2) when hosts, both livestock and wild ungulates, are numerous and widely spread, allowing high rates of contact between hosts and vectors.

The strong correlation between maximum temperature and trypanosome prevalence reported in this study confirms the relationship between environmental temperature and trypanosome development (Walshe et al. 2009 ). Environmental temperature regulates the switching of the trypanosomes from the bloodstream form that has entered the fly and their multiplication in the tsetse fly midgut (procyclic form) before proceeding to salivary glands ready to infect a host (Akoda et al. 2009 ). It appears from in vitro experiments that the process of differentiation of trypanosomes from the blood stream form to the procyclic form initiates and continues the multiplication when the temperature is between 37° to 27° C / 26° C (Brown et al. 1973 , Bienen et al. 1980 , Milne et al. 1998 , Li et al. 2003 ). Other laboratory experiments concluded that 25° to 28° C was the optimum temperature for growth of trypanosomes but slow growth occurs up to 37° C (Cross and Manning 1973 ). Generally, high temperatures shorten the duration of trypanosome development cycles within the tsetse fly (Moore et al. 2011 ). We reported infections between 26° C to 36° C with the highest infection rates at around 31° C. There could be two reasons for this observation; first, there is the rapid differentiation and proliferation of trypanosome species at temperatures between 26° C and 37° C as suggested by laboratory studies; second, nutritional stress could affect these patterns where high temperatures induced quick digestion of blood meals in the tsetse fly leading to more frequent feeding events, and hence increased risk of infection. Other exo‐endogenous factors might also play an important role, such as quality of blood source, type of tsetse fly midgut enzymes, and quality of parasite surface coat, though they were beyond the scope of this study (Kubi 2006, Akoda et al. 2009 , Geiger et al. 2015 ).

Peaks in trypanosome prevalence lagged behind tsetse fly abundance peaks and confirmed previous findings by others (Rogers 1988 ). This observation may be partly because tsetse flies catch infections from hosts, and we suspect the flies to be susceptible to low immunity during low fly abundance and vice versa, but we did not have data to test this. Since tsetse fly infection rates depend on prevalence of trypanosomes in the vertebrate host populations, the observed general inverse relationship between tsetse fly relative abundance and trypanosomes prevalence implies that there is a risk of trypanosomiasis infection by vertebrate hosts regardless of tsetse fly abundance. Nonetheless, very low vector abundance lowers the prevalence as shown in January, indicating that a threshold of tsetse fly abundance is important for trypanosome infections to occur.

In summary, this study highlights seasonal patterns of tsetse fly burden and trypanosome prevalence. This information can inform stakeholders about the months of the highest risk of trypanosomiasis infection in the Maasai Steppe, where this information is limited or unavailable. Also, the established temporal patterns of vector and parasites together with climate relationships can provide fundamental information needed when developing predictive models of the temporal dynamics of the relative abundance of individual tsetse fly species and their infection rates. Nonetheless, it is important to recognize the possible role of other factors associated with changes in climate and how they may affect trypanosomiasis dynamics. Furthermore, the experimental thermal tolerance of G. swynnertoni and G. m. morsitans under field conditions is recommended in order to explain the variation of temperature effects to this species reported in this study.

No comments

Добавить комментарий

Your e-mail will not be published. All fields are required.