Tsetse flies belong to the single genus Glossina, in the family Glossinidae of the order Diptera. They are found only in sub-Saharan Africa, Yemen, and Saudi Arabia, infesting 38 countries and occupying about 11 million square kilometers. As cyclical vectors of protozoan parasites of the genus Trypanosoma, they are of major economic and biological importance. Trypanosomosis (sleeping sickness in humans) is a major constraint to livestock production and is a threat to millions of people in Africa. Within the genus, 31 species and subspecies of tsetse have been identified. Fossil Glossina, found in Florissant shales in Colorado, date back to the Oligocene, indicating a wider original distribution..
The genus is split into three subgenera or groups: Austenina (fusca group), Nemorhina (palpalis group), and Glossina (morsi-tans group). The forest-dwelling f usca group are the most primitive. Classification is based largely on morphology of the genitalia; however, the three groups also differ ecologically. Differences in the cuticular alkanes of their sex pheromones can also be used for classification and for estimating “genetic” distance between species.
LIFE CYCLE AND REPRODUCTION
An adult female produces one egg at a time, from which a first instar hatches in the uterus, where it is supplied with nutrients. After a period of development and molting in the uterus, a third instar is deposited on the ground. One full-grown larva is produced every 9 or 10 days, which then pupates in light or sandy soil. The adult emerges after a puparial period of around 30 days, depending on the ambient temperature. This process, termed adenotrophic viviparity, results in a very low rate of reproduction (Fig. 1).
FIGURE 1 An adult tsetse fly in the process of feeding.
Spermatogenesis occurs during the puparial stage of males and ceases after eclosion; maturation of adult males takes approximately 3 days. Substances synthesized in the male accessory glands are transferred to the female hemolymph during copulation and stimulate receptivity of female tsetse. Unfed females are unwilling to copulate, and receptivity declines with age or after mating. Most female tsetse are successfully inseminated even at very low population densities and need to mate only once in their lifetime, although multiple mating does occur. Male flies seem to rely on visual attraction to find mates ( Fig. 2 ).
Larval Nutrition and Development
In pregnant females, proteins synthesized by highly efficient “milk” glands provide nourishment to the developing larva. Milk is secreted cyclically, with a peak in early pregnancy, which continues from the time of larval hatch until parturition. Some components of the milk probably originate from fat-body secretions, and the milk contains relatively constant proportions of protein and lipid plus phospholipids, cholesterol, triglycerides, and tyrosine. The latter may be necessary for tanning larval and adult cuticles. At eclosion of the first instar, the milk is rich in acidic lipids, which are later replaced by proteins.
The anterodorsal mouth of the first instar is blocked by a chiti-nized median tooth. This “egg-tooth,” lost with the skin of the first instar, is used to puncture the eggshell (chorion), which then splits along the dorsal side, allowing the first instar to hatch.
Larviposition and Pupariation
Unlike most dipteran larvae, which commit to metamorphosis early in the third instar, the third instar tsetse may pupariate only when fully grown. Shortly after larviposition, larvae exhibit a strong photonegative response and attempt to burrow. Uric acid from the
FIGURE 2 Life cycle of the tsetse fly.
anus then spreads over the larva and may protect it from predation. Periodicity of larviposition is affected by temperature, and the tsetse may respond to seasonal temperature cycles such that larviposition occurs in the most favorable conditions. The female fly becomes very active a few hours before parturition, presumably while searching for a suitable larviposition site. Both tactile and visual responses have a role in breeding site location. Females tend to choose darker resting sites as pregnancy progresses and when temperatures increase. The interlarval period depends on maternal nutrition and temperature conditions and varies according to species, localities, and seasons, from 20 to 90 days. Survival of puparia is dependent on relative humidity, which, together with temperature, influences selection of breeding and larviposition sites. Survival of emerging adults depends on the existence of sufficient fat reserves to support them until their first blood meal.
The adult emerges from the puparium using the “ptilinum,” a saclike structure folded in a frontal cavity of the head. Rhythmic contractions of thoracic and abdominal muscles help the adult to break out of the puparium and reach the soil surface. Females emerge 1-2 days before males. Immediately after eclosion, the wings, abdomen, and thorax are highly compressed, although the legs are already full sized. The newly emerged fly has to expand by about 90%, using the cibarial pump to suck air into the gut and must then feed to complete growth of the flight muscles and endocuticle.
BLOOD MEAL DIGESTION AND UTILIZATION
During feeding, tsetse discharge saliva containing a powerful anticoagulant antithrombin enzyme. The quantity of saliva secreted increases as tsetse become hungrier and may contain trypanosomes in infected flies. Two platelet aggregation inhibitors identified in saliva may also have immunosuppressive, anti-inflammatory properties. Digestion of blood proteins takes place in the posterior section of the midgut and involves six enzymes, which convert proteins to peptides and free amino acids.
Water Balance and Excretion
After feeding, the increased weight of the fly hampers flight and increases vulnerability to predators. There is therefore a rapid reduction of the water content of the blood meal from 79% to about 55% within 3 h after feeding. A blood meal in early pregnancy provides females with nutrients that are stored for larval development in late pregnancy.
Pheromones and Hormones
Sex recognition pheromones of tsetse are saturated, methyl-branched hydrocarbons found in cuticular waxes and are apparently unique to each tsetse species. They trigger mating behavior in males at ultraclose range or upon contact with baited decoys. The pherom-one is released from the cuticle via unicellular glands on the legs and spreads over the external body surface by diffusion and grooming behavior. It is present on the pharate adult female about 2 days before emergence from the puparium and remains throughout life. A larval pheromone may be released from the anal orifice of tsetse larvae. The pheromone seems to affect choice of larviposition site, resulting in aggregation of larvae in breeding sites.
The energy required for flight is produced from proline and is sustained by utilization of lipid reserves. Proline is depleted during flight and cannot rapidly be replenished; therefore, tsetse are limited to short periods of high-speed flight, at between 6.5 and 7.5 msec-1 (0.4km min-1) in open country. Proline is synthesized from lipids in fat bodies and oxidized to alanine in the muscles under hormonal control. Daily flight duration in male flies is about 15min in the hot season and more than twice as long in the cold season. If fat reserves fall to about 6% of total dry body mass, the fly may die of starvation; such nutritional stress also results in production of small puparia.
Vision and Sense Organs
The compound eyes of both sexes are similar, with a specialized zone of greater visual acuity in the forward-pointing region. Males, however, have a wider region of binocular overlap, which may enhance detection of females. The visual function is consistent with fast flight, detection of drift due to low wind speeds, mating chases, and discrimination of cryptic hosts at high light intensities.
For mating behavior to be initiated, both mechano- and chemore-ceptors must be stimulated. The chemoreceptors are thought to be basiconic sensilla on the tibiae and tarsi. Other hair-shaped chemo-and mechanoreceptor sensilla are found on the front side of the wings.
DISTRIBUTION AND HABITATS
The general distribution of tsetse flies, determined principally by climate, is influenced by altitude, vegetation, and the presence of suitable host animals. The limit of distribution is closely correlated with the “tropical savanna (summer rain) climate,” which follows the >508mm annual rainfall line, whereas the “tropical rain forest (equatorial) climate” controls the habitats of the fusca and palpalis groups. The surrounding savannas are the habitat of the morsitans group. The southern limits of Glossina distribution in Africa lie north of a line drawn from Benguela, in Angola, to Durban, in South Africa. The northern limits are roughly a line from Dakar in Senegal across to Ethiopia and Mogadishu in Somalia on the east coast.
The distribution and abundance of morsitans group tsetse often correspond to those of wild animals. In West Africa, G. longipalpis inhabits forest islands with plenty of shade and moves to forest edges only during seasons of high rainfall. G. austeni is confined to the coastal zone of East Africa, from Somalia to South Africa, including the island of Zanzibar until it was eradicated from the island in 1997.
Species of the palpalis group occur in drainage systems leading to the Atlantic Ocean or to the Mediterranean, not those draining into the Indian Ocean. G. palpalis is found close to water, in gallery forests, and it cannot tolerate the wide range of climatic conditions occurring in the savanna belt, where it is restricted to watercourses or ” forest islands. ”
Species of the fusca group are mostly found in forests of West and central Africa. Humans are likely to have an adverse effect on fusca group tsetse populations through forest clearing and hunting.
Human activity has an important influence on tsetse distribution and abundance; humans can scare away or kill potential hosts, or destroy the vegetation forming the flies’ habitat through agricultural or other development. Most African countries, particularly those in tsetse-infested areas, have low human population densities; however, Nigeria, Africa’s most densely populated country, has a population of 89-100 million, equivalent to 108 people km~2. Nigeria developed rapidly, and as a result, up to two-thirds of the potential G. m. sub-morsitans population of Nigeria may have been suppressed. G. m. submorsitans occurs in areas with human population densities ranging from 0 to 15 km~2, occasionally in areas of 15-40 km~2, but never when the population exceeds 40 km~2. Most sub-Saharan African countries have densities below 40 km ~2. Flies of the palpalis group, particularly G. tachinoides, are much less affected by human settlement, possibly because they are able to adapt from a preference for feeding on wild mammals and reptiles, to feeding mainly on humans and their domestic animals. Tsetse flies can adapt to peridomestic habitats, and G. tachinoides commonly follows domestic pigs in villages and utilizes human-made larviposition sites such as clumps of oil palm, cola nut, and banana trees.
HOSTS AND FEEDING BEHAVIOR
Potential hosts are recognized from visual and olfactory stimuli, which, together with mechanical stimulation, activate tsetse and initiate host-oriented responses. Approach to a stationary host is by upwind flight, modulated by olfactory stimuli, flight speed significantly reduced when a fly enters an odor plume. Final orientation toward the host is visual. Heat stimulation after a fly has landed on the host may then cause a probing response and subsequent feeding. Both endogenous and exogenous factors influence host-seeking behavior. Endogenous factors include a circadian rhythm of activity, as well as the level of starvation, age, sex, and pregnancy status of the flies. Exogenous factors include temperature, vapor pressure deficit, and visual, mechanical, and olfactory stimuli.
Odor attractants are of three types: (1) those associated with animal breath, such as acetone, octenol, and CO2; (2) those associated with urine, such as the phenols; and (3) those associated with skin secretions, such as sebum.
As a tsetse fly lands on a host, the labium is enclosed between the palps. Heat is the prime stimulus leading to probing and subsequent feeding by tsetse. Responsiveness increases with hunger until the final stages of starvation. As the fly starts to probe, the labium moves from the palps to an angle of 90° to the skin. While the labella rests on the skin, the teeth on the inner surface are everted and penetrate it. At the same time, saliva is excreted from the hypopharynx. Normally, the labellar teeth lacerate capillaries, resulting in a hemorrhage, which is sucked into the labrum. When the fly stops sucking, a small pool of blood forms. If blood is not found, the fly withdraws the labium partially and makes a new penetration. Unmated female tsetse take smaller meals than mated ones, because either the gut cannot be distended to the same extent or of lower metabolic demands.
Host Effects and Preferences
Nonrandom feeding patterns may result from responses of hosts to tsetse attack, such as tail-flicking or skin-rippling reactions. Although normally unable to discriminate between potential hosts at long range, the upright habit of humans, and lactic acid secretions in the skin, result in visual and olfactory repellency, including inhibited landing responses. Zebras may be protected from being fed on by biting flies, including tsetse, by their striped pattern. Tsetse are less attracted to stripes than to solid colors and seem to avoid horizontally striped objects. Zebras usually have a combination of vertical and horizontal stripes, but the horizontal stripes are on the lower part of the animal, where tsetse normally feeds.
There is a close correspondence between ecological niches of common hosts and tsetse habitats, and overlapping habitats are important in determining feeding patterns, in addition to behavioral characteristics of the host.
Tsetse will change to alternate hosts if their usual host(s) become unavailable. For example, G. longipennis, which once fed mainly on rhinoceros, adapted to feeding on other large animals as rhinoceros became scarce. Tsetse select host species rather than simply feeding on the most common available host.
Wild pigs (Suidae), particularly warthogs, are important hosts for morsitans group tsetse; however, the feeding efficiency of tsetse visiting a single warthog could be as low as 12-18%. Between 26% and 31% of tsetse landed on the head region of live adult warthogs, apparently because of a visual response to a dark patch produced by the preorbital glands of mature warthogs. In East Africa, Suidae can form about half the food supply of morsitans group tsetse in areas with wildlife. Most other feeds are from ruminants, particularly bushbuck.
The palpalis group, generally inhabiting riverine or lacustrine vegetation, have a close ecological association with important hosts such as crocodiles and monitor lizards. They are, however, opportunistic feeders, and many mammals, including humans and reptiles, that enter their habitat may be fed upon. In peridomestic situations in the Guinean “forest savanna mosaic” region of West Africa, domestic pigs are an important source of blood meals for G. palpalis and G. tachinoides; wild Suidae, humans, and small ruminants are the next most common hosts. In the forest zone of Cote d’Ivoire, the feeding habits of G. palpalis vary with the availability of wild hosts and activities of humans. Around villages, nearly all feeds are from domestic pigs, whereas in plantations, more feeds are from humans. In human trypanosomosis foci outside the edges of villages, pal-palis group flies may feed equally on humans and antelopes, especially bushbuck. G. tachinoides can survive in close association with humans in the virtual absence of wild mammals and reptiles. It will readily feed on domestic pigs or cattle, but feeds have rarely been identified from domestic sheep and goats even when these were common.
Tsetse rest for most of the day, making use of sites that provide protection from extreme temperatures and from predators. The sites may also provide vantage points from which to seek hosts, and their location can change according to time of day, season, and hunger stage of the fly. Recently engorged tsetse fly poorly and rest low down on tree trunks; the height above ground level of the resting site increases in relation to the fly’s nutritional state.
During the daytime, high temperatures of exposed resting sites result in tsetse moving to rest in sites such as the boles of large trees, where they squeeze into the fissures of the bark, at heights generally less than 0.3 m from the ground. Otherwise, flies hide in rot holes, often quite high up, in big tree trunks.
At dusk, tsetse flies generally move upward, to spend the night on leaves or small twigs, possibly to avoid predators. The choice of nocturnal resting sites is influenced by seasonal effects on the physical condition of the vegetation, such as leaf fall. A reverse migration seems to take place about an hour after sunrise, possibly in response to light intensity falling below a critical level. This rapid migration may occur at different temperatures in different seasons.
The diurnal pattern of tsetse activity may differ between species and according to sex, hunger, pregnancy, and nutritional state of the population. The pattern may also differ for the same species of tsetse in different localities. Natural activity of tsetse flies can be environmentally stimulated or spontaneous, but it has an underlying endogenous circadian rhythm modified by temperature. Intensive activity occurs for periods of less than a minute. The morsitans group tsetse are mostly active early in the morning and/or late in the afternoon.
DEVELOPMENT OF TRYPANOSOMES IN TSETSE
Fly species differ in their capacity to transmit trypanosomes, and individual fly genotypes also vary and affect susceptibility to trypanosome infection. For example, salmon mutants of G. m. mor-sitans appear to be better vectors of trypanosomes, perhaps owing the metabolism of tryptophan, which is essential for trypanosomes. Metabolism in salmon mutants is affected so that tryptophan accumulates, predisposing the fly to trypanosome infection. Flies of the morsitans group are mostly good vectors of all trypanosome species. Species of the palpalis group seem to be poor vectors of Trypanosoma congolense but efficient vectors of some stocks of T. vivax and can be important vectors of human infective trypano-somes. Species of the fusca group can be effective vectors of T. congolense and T. vivax but poor vectors of Trypanozoon trypanosomes. Trypanosome infection rates in tsetse are generally low and can depend on the age of the fly at the time of the infective feed. Wild male tsetse can achieve a life span of almost 5 months, but this is probably unusual. Teneral flies taking an infective blood meal on the first day of life, or soon after emergence, are the most easily infected. The prevalence of mature infections appears to increase with age for Nannomonas and Duttonella group trypanosomes in many tsetse species. Susceptibility to trypanosome infection may be age dependent, but the actual age is less critical than whether the fly has taken a previous, uninfected feed.
Trypanosomes were first known to infect humans after being detected in the blood of a steamboat captain in the Gambia in 1901. The parasite was identified and named Trypanosoma gambiense. Sleeping sickness currently occurs in about 200 distinct disease foci in Africa. These foci place between 35 and 55 million people at risk, although only about 3 million of them are under surveillance and relatively few new cases are diagnosed annually. Human sleeping sickness was largely under control in the 1960s, but its incidence then increased, and in 1994 it was estimated that there were 150,000 cases in the Democratic Republic of the Congo. Other estimates suggest that around 300,000 people are infected in Africa, many of whom will die because of lack of treatment (Fig. 3).
FIGURE 3 Distribution of human sleeping sickness foci in Africa. Dashed line indicates boundary between foci of T. b. gambiense (west) and T. b. rhodesiense (east). Arrows denote the probable direction of spread of the two types.
Identity and Origins of the Parasite
The two parasites causing human sleeping sickness, T. brucei gambiense and T. b. rhodesiense, are morphologically indistinguishable, and early diagnosis was based on clinical signs of the disease.
Sleeping sickness caused by T. b. rhodesiense is an acute disease, with death occurring after only a few months, whereas with T. b. gambi-ense, death may not occur for several years. T. b. gambiense occurs in West and central Africa and is transmitted predominantly by palpalis group tsetse, while T. b. rhodesiense occurs in southern and East Africa and is transmitted by morsitans group flies.
Although generally accepted that sleeping sickness trypanosomes are derived from T. b. brucei (infecting wild and domestic animals and morphologically indistinguishable from the human infective parasites), there are several theories on the evolution of species infecting humans. First, the two types of sleeping sickness may be caused by the same parasite, the disease simply being more acute when the trypanosome becomes adapted to certain species of wild animals. Chronic disease occurs when the trypanosome is adapted to humans as its principal host. Second, T. b. rhodesiense may have evolved in the late Miocene or Pliocene, when hominids were exposed to Trypanozoon trypanosomes in their savanna habitat. Subsequently, T. b. gambiense may have evolved from T. b. rhodesiense when homi-nids invaded forests and became hosts of palpalis group tsetse. Third, the two subspecies may have evolved independently from T. b. brucei or a common ancestral species. A fourth theory is that after T. b. gam-biense spread to savanna areas of southeast Africa, T. b. rhodesiense evolved from it and subsequently spread northward. An alternative theory is that T. b. brucei gave rise first to T. b. rhodesiense and then to T. b. gambiense and that the two diseases evolved in humans from an animal infection to an anthropozoonosis (T. b. rhodesiense), and then to a pure anthroponosis (T. b. gambiense).
Molecular characterization of T. b. rhodesiense suggests that it had polyphyletic origins. It is widely accepted that T. b. gambiense sleeping sickness spreads from West to central Africa. Areas of West Africa where sleeping sickness occurred and the Nile basin are ecologically similar, with a continuous corridor of Sudano-Guinean climate and vegetation. T. b. rhodesiense then would have spread northward either from Central Africa through East Africa or from the Zambezi valley in Zambia, increasing in virulence the further north it progressed. Contradictory evidence from biochemical and molecular characterizations suggests that strains of T. b. rhodesiense from Zambia, Kenya, and Uganda have independent origins.
Until recently, one of the difficulties in proving the existence of reservoir hosts for T. b. rhodesiense and T. b. gambiense arose from problems in subspecies identification. It is now possible to use molecular genetic and biochemical methods to identify human infective forms. Consequently, there is now evidence of a range of wild and domestic animal host reservoirs. The latter are likely to be less important in the epidemiology of sleeping sickness due to T. b. gambiense than for T. b. rhodesiense. Many animals can be experimentally infected with either subspecies, and wild and domestic animals can maintain the “rhodesiense” disease endemically.