Pathogenicity of trypanosomes in relation to 1 drug sensitivity: comparative studies between a drug-sensitive and drug-resistant Trypanosoma congolense strain in murine-and bovine model

Background: Indiscriminate exposure of Trypanosoma congolense and other trypanosomes to trypanocides have led to emergence of resistant strains, which increasingly undermine the control efforts against tsetse transmitted animal trypanosomosis. Despite the increasing trend of resistance, no studies have conducted to assess whether or not drug sensitivity affects the pathogenicity of T. congolense and other predominant species. Methods: We compared the pathogenicity of two strains of trypanosomes: drug-sensitive Trypanosoma congolense -Mikese and drug-resistant T. congolense Mbagala. These strains were isolated from cattle at Mikese, Morogoro region and Mbagala, Dar es Salaam region, Tanzania, respectively. Experimental mice and cattle were infected with either tryapnosomes and monitored for rectal temperature, prepatent period, parasitemia, packed cell volume (PCV), survival and clinical manifestations. Drug sensitivity status of either strains was re-conrmed before pathogenicity testing. Results: Mean rectal temperature was higher in mice infected with T. congolense -sensitive strain (p=0.049, 95% CI= 0.003-1.13). Mean prepatent period of resistant strain in mice and cattle was shorter than that of sensitive strain. The mean level of parasitemia was signicantly higher in resistant- (7.5±0.8) than in sensitive-strain (6.5±0.8) (p<0.001, 95% CI= -1.28 to -0.68). Mice infected with resistant strain were relatively dull and lethargic compared to those infected with sensitive strain. The decline PCV was higher in cattle infected with sensitive- strain than resistant-strain (p=0.041, 95% CI, -6.97 to -0.17). Conclusion: Pathogenicity of the two Trypanosoma congolense strains varied signicantly across host species. The resistant strain was highly pathogenic in mice and less so in cattle. Contrarily, the sensitive strain was highly pathogenic to cattle and less so to mice. As such, this study emphasizes variations on the pathways by which different trypanosomes act upon the host; thus warranting subsequent studies using large number of experimental animals, preferentially cattle, in view of reecting the eld situation.


Introduction
Tsetse-transmitted Animal African Trypanosomosis (AAT) is the major constraint to livestock production in sub-Saharan Africa (SSA). It has direct impacts on livestock production, livestock management and human settlement; and indirect impacts on agriculture and human welfare [1]. Its annual estimated direct and indirect losses run into billions of dollars [2]. The disease is endemic in more than 37 African countries and its geographical map may be increasing consequent to land-use and climate changes. Extracellular protozoan parasites belonging to the genus Trypanosoma are causative agents of the disease. Main trypanosomes responsible for the disease cattle include T. congolense, T. vivax, and to a lesser extent T. brucei. Of these, T. congolense is the most prevalent and pathogenic species [3].
A range of domestic and wild animals are affected by trypanosomosis, however, their susceptibility to different trypanosomes varies among them. Cattle are the most affected animals and experience the greatest disease burden. The variation in susceptibility is determined by breed, age, behavior, previous exposure and health status of host species as well as species and/or strains of trypanosomes [4]. These factors also dictate the pathogenicity of trypanosomes [5].
The most consistent clinical manifestations of trypanosomosis in cattle include intermittent fever, anaemia and weight loss [6]. Death in some animals may occur often as a result of congestive heart failure due to anaemia, myocarditis and circulatory disturbances. There are no manifestations and/or signs that are pathognomonic to trypanosomosis. In susceptible cattle, however, the development of anemia is regarded as a cardinal sign of trypanosomosis and is used as a fundamental diagnostic feature for the disease [7].
The control and treatment of AAT, particularly in resource poor communities of SSA, exclusively relay on the use of trypanocidal drugs. Vaccine development has been hindered by the ability of trypanosomes to evade host immune response via an elaborate mechanism combining antigenic variation and immunosuppression [8]. Tsetse control has proven to be relatively complex and expensive [9].
Diminazene aceturate (DM) and isometamidium chloride (ISM) are the most frequently and widely used trypanocidal drugs. They have been in use for nearly half a century [10,11] and have considerably reduced disease associated losses particularly in areas where they have been deployed regularly and appropriately. However, the effectiveness of these drugs is increasingly undermined by the emergence of resistance particularly in predominant Trypanosoma species [12]. Resistance to trypanocides has been reported in over 18 African countries, including Tanzania [13][14][15][16][17]. Even more worrying are the increasing reports of multiple drug resistance [17,18].
Despite the increasing problem of drug resistance, whether or not such trait affects the pathogenicity of T. congolense and other frequent trypanosomes, negatively or positively, remains indeterminate. Pathogenicity determines progression of a pathogen in its host, thus determines how a host and its parasite interact in a given area. Several studies have shown profound variation in virulence amongst subgroups of trypanosomes [19][20][21] and strains [5,22] of Trypanosoma species. Besides, studies have shown loss of virulence and/or tness in drug resistant trypanosomes [23]. This study sought to determine the correlation between drug sensitivity and pathogenicity of putatively drug sensitive Trypanosoma congolense-Mikese and drug resistant T. congolense Mbagala. These strains were isolated from cattle at Mikese, Morogoro region and Mbagala, Dar es Salaam region respectively.

Experimental animals
Two types of animals were used in this study: swiss albino mice and indigenous shorthorn zebu cattle. For the case of mice, we used 8-12 weeks old males, weighing 20-30g, obtained from Small Animals Breeding Unit, College of Veterinary Medicine and Biomedical Sciences, Sokoine University of Agriculture, Tanzania. These mice were labeled using picric acid per procedures described in Lumsden et al. [24]; and maintained under room temperature within a well-ventilated y-proof experimental house. Wood chippings were used as bedding materials and were frequently changed to ensure dry condition. The mice were fed on broiler mash commercial preparation (in lieu of speci c mouse feed formula). Water was provided ad libitum. We used shorthorn zebu steers, 6-8 months old, weighing 120-140kg purchased from the nearby villages known to be tsetse free. They were ear-tagged and maintained under room temperature in a y proof pen at Sokoine University of Agriculture (SUA). The steers were acclimatized for four months, during which they were also screened for worms, trypanosomes and other haemoparasites, after which they were used for the experiments. The steers were regularly administered with Levamisole at the dose of 1ml per 10kg body weight and imidocarb at the dose of 3mg/kg body weight to prevent them from worms and other haemoparasites respectively. Packed cell volume (PCV) was also examined and recorded as baseline data. The animals were fed on cut grass and supplemented with hay and concentrates. Water and a mineral lick (farmers superlick®) were provided ad libitum throughout the study.

Trypanosomes
Two strains of trypanosomes: Trypanosoma congolense-Mikese (drug sensitive) and T. congolense Mbagala (drug-resistant strain), were used in this study. These strains were isolated from cattle at Mikese, Morogoro region and Mbagala, Dar es Salaam region respectively. Our recent sequencing and phylogenetic analysis revealed that the two strains are genetically distinct The drug-resistant T. congolense Mbagala and drug-sensitive T. congolense Mikese were identi ed as T. congolense savannah and T. congolense Kili respectively (Ngumbi and Mnyone, Unpublished). These strains are maintained through serial passages in mice at the Small Animals Breeding Unit, College of Veterinary Medicine and Biomedical Sciences, SUA, Morogoro, Tanzania.

Pathogenicity testing in Swiss albino mice
The mice were divided into 2 treatment groups and one control group, 5 mice each. In treatment group one, the cohort of mice were inoculated with the drug sensitive strain; referred herein as the T. congolensesensitive group (TCSG). In treatment group two, the cohort of mice were inoculated with drug resistant strain; referred herein as the T. congolense-resistant group (TCRG). The last cohort of mice were inoculated with phosphate buffered saline with glucose (PBSG); referred herein as the control group (CG).
The inoculation of both strain (each 2 ml of 5×10 6 trypanosomes/ml) and PBSG was done intraperitoneally. These mice were monitored for pre-patent period (time-lapse post-infection to parasite detection in blood), parasitemia and survival for over 30 days post infection. Parasitaemia was monitored from day 2 to 14 post infection. These parameters were done via microscopic examination of wet tail blood lms [25].

Pathogenicity testing in indigenous shorthorn zebu steers
Three steers were used in this experiment and each received a different treatment as follows: sensitive T. congolense strain (BIS1), resistant strain (BIS2) and PBSG (BIS3). The last steer served as the control. Similar to the case of experimental mice above, the drug sensitive and resistant T. congolense strains inoculated to steers were obtained from the donor mice. The administration of both strains (each 5 ml of 5×10 6 trypanosomes/ml) and PBSG was done via the intravenous route [25]. Afterwards, all these steers were all monitored for prepatent period (time-lapse post-infection to parasite detection in blood), parasitemia, anaemia, clinical signs and survival. These parameters were assessed through wet blood lms and buffy coat microscopic examination, packed cell volume (PCV) and visual examination of steers. Their rectal temperature was also monitored on daily basis. The monitoring of all these parameters was done three times a week. Blood samples were collected from the jugular vein into EDTA vacutainer tubes. The wet blood lms and buffy coat were examined for trypanosomes with the aid of compound microscope at 400× magni cation. Anaemia was assessed through estimation of PCV using micro-haematocrit centrifugation technique. Similar to mice, the levels of infection were categorized as acute, sub-acute and chronic [19].

Data analysis
The data collected from all experiments were entered and managed in excel. All data analysis was conducted using SPSS statistical software (in SPSS version 16). Student's t-test was used to compare treatment and control animals (mice and steers) in terms of their mean prepatent period, parasitaemia, anaemia and proportion surviving.

Pathogenicity in Swiss albino mice
The mean prepatent period did not differ between mice infected with drug sensitive strain days (4.6±1.5) and drug resistant strain (2.8±1.3 days) (P=0.079). The parasitaemia in mice infected with drug resistant strain (7.5±0.8 trypanosomes/ml) was signi cantly higher than in mice infected with drug sensitive strain (6.5±0.8 trypanosomes/ml) (P<0.001, 95% CI= -1.28 to -0.68). Parasitaemia increased with time in both strains. However, in mice infected with resistant strain, parasitemia stabilized at 1×10 7.5-8.1 trypanosomes/ml from day 8 post-infection to the end of the monitoring period ( Figure 1). The mean rectal temperature was signi cantly higher in mice infected with sensitive strain (37.6±0.6 o C) than in mice infected with resistant strain (37.0±1.3 o C) (P=0.049, 95% CI= 0.003-1.13). Furthermore, the rectal temperature mice infected drug resistant strain declined to 36.5 o C at day 6 and stabilized at day 8 postinfection correlating to stabilization of parasitaemia. The mice infected with drug sensitive strain experienced episodes of temperature elevations throughout the monitoring period ( Figure 2). Moreover, the mice infected with resistant strain appeared relatively dull and lethargic compared to the mice infected with sensitive strain. The latter cohort of mice remained active over the entire monitoring period. Mortalities were recorded between day 9 and 30 for mice infected with resistant strain, whereas mice infected with sensitive strain survived for more than 30 days post infection.

Pathogenicity in indigenous shorthorn zebu steers
Whereas the mean prepatent period for the steer infected with drug sensitive strain was 9 days that of the steer infected with drug resistant strain was 7 days. Parasitaemia in the steer with resistant strain increased few days post-infection and thereafter decreased to 1×10 5.4 trypanosomes/ml. Parasitemia in the steer with sensitive strain increased to 1×10 7.5 trypanosomes/ml on day 26; and eventually died in the course of treatment, 41 days post infection. The steer had been cleared of trypanosomes by the time it died. Overall the PCV in both of infected steers declined signi cantly compared to that of uninfected steer (control). The mean PCV was 18.1±2.2 for steers infected with sensitive strain and 21.7±3.5 for steers infected with resistant strain. However, the decline in PCV was signi cantly higher in the steer infected with sensitive strain than in the steer with resistant strain (P=0.041; 95% CI, -6.97 to -0.17) (Figure 3). In terms of clinical signs, whereas the steer infected with resistant remained asymptomatic throughout, the steer infected with sensitive strain presented with poor body condition, dullness and reluctance to eat.

Discussion
In this study, we aimed to assess whether or not the resistance of T. congolense strain to trypanocides affects the development of infection in susceptible hosts. The slightly shorter prepatent period of drugresistant T. congolense Mbagala in mice and steer relates with the observations of Tesfaye and coworkers [26] where resistant and sensitive isolates of T. congolense were compared in goats. Unlike these two studies, however, most reports revealed shorter prepatent period in drug sensitive strains [27]. This theory coincides with the proliferation and differentiation phenomenon of actively dividing slender form of trypanosomes, T. brucei in particular [27]. Findings of the present study therefore portray greater ability of the drug-resistant T. congolense Mbagala to multiply and establish in new hosts as relative to drugsensitive T. congolense Mikese. This is the reason why parasitemia in mice and cattle infected with resistant strain appeared relatively faster compared to the sensitive strain. This unpopular short prepatent period and high rate of parasitemia observed in T. congolense Mbagala strain could be emanating from selection pressure created by extensive syringe passage, which often tend to select for parasite with higher replication rates. Similar scenario may be happening to the eld population of trypanosomes in cases of intense selection pressure; thus emphasizing the need for collecting and testing fresh resistant isolate from the eld where the current one was originally collected. Furthermore, the high multiplication rate in the drug resistant strain could also be explained by the ability of resistant trypanosomes to survive a die-off phenomenon owing to host's immunological responses and possible rapid expression of variable surface glycoprotein (VSG) [28].
Although done de nitely for the purpose of re-examining whether or not the test strains were still retaining their original drug sensitivity status, our results emphasize maintenance of such traits despite undergoing serial passage in laboratory white mice [29]. This observation, however, warrants further studies to ascertain how long such traits are retained particularly in cases of short cycles of repeated serial passages.
Worth noting, the initially high level of parasitaemia observed in cattle infected with resistant strain, declined over time post infection. Similar pattern was also observed in the trend of packed cell volume (PCV), which was measured as a proxy for anaemia. Such patterns varied prominently with observations made when mice were infected with such strain; suggesting the role of host immune response in limiting the increase of parasitaemia and anaemia. This phenomenon is among other factors that bene t trypanotolerant breeds of cattle. Naessens [30] de ned trypanotolerance in bovine as the better capacity to control anaemia throughout the infection and eventually control parasite numbers in the blood during the later stages of infection. Low parasitaemia, moderate anaemia and non-disease state was also observed in this study in steer infected with resistant strain of T. congolense. Marcotty et al. [31] explained more parasite detection at early stages of infection even before anaemia development but low parasitaemia with marked anaemia at later stages of infections. The decline of parasitaemia and anaemia in cattle with resistant strain was associated with low rate of switching of variant surface glycoprotein (VSG) coat of trypanosomes attributed to serial syringe passaged infection [32].
Moreover, the mean PCV steer infected with the sensitive strain were signi cantly lower than that of steer infected with the resistant stock (p=0.04). This observation a rmed that there was a difference in the level of anaemia exerted by either strain; and emphasized anaemia is one of the major clinical manifestations of Trypanosoma infections. In this case anaemia is caused by mechanical injury of erythrocytes due to lashing movement of trypanosomes during parasitaemia [33]. Mbaya et al. [34] observed severe anaemia which was associated with high parasitaemia in gazelles concurrently infected with T. congolense and Haemonchus contortus. As expected, this and other similar ndings corroborate with ndings of the present studies in that the severity of anaemia and its correlation with parasitemia.
Scanty or absence of parasitaemia has been a characteristic feature in chronic form of T. congolense infections in cattle. This is normally accompanied by severe illness and death. Contrary was observed in this study, that parasitaemia in steer infected with T. congolense resistant strain declined over time resulting into asymptomatic host. However, the situation was different in steer infected with T. congolense-sensitive strain which demonstrated signi cantly low PCV, relatively higher parasitaemia, which resulted into development of the disease and eventually death of the steer within 31 days of monitoring. The animal died in the course of treatment. Mortalities in cattle after 4-7 weeks after infection with T. congolense strains were also reported elsewhere [20]. Based on the results of this study, the T. congolense-sensitive strain in cattle seemed more pathogenic that the resistant strain due to the fact that it eventually caused more severe anaemia, clinical manifestation and nally killed the animal.
The characteristic uctuation in the rectal body temperature was also observed in the present study. There was increase in rectal body temperature in both groups in the second and third day post infection for sensitive and resistant strains infected mice. Despite both groups showing rise in rectal temperature, the rise was signi cantly higher (P=0.049) in mice infected with sensitive strains. The uctuation in rectal temperature was observed throughout the parasitaemic phase and these observations were consistent with those reported by Mbaya et al. [35] following T. brucei infection in gazelles.
In the course of this study it was observed that mice infected with T. congolense-resistant strain appeared dull with reduced activities in a way that made them easy to handle as compared to those infected with T.
congolense-sensitive strain. Mortalities were observed in T. congolense-resistant strain as early as nine days post-infection and hardly struggled to survive for 30 days since inoculation. In view of the observations mentioned above, T. congolense-resistant and sensitive strain were categorized as highly pathogenic and moderately pathogenic in mice; and this was in accordance to the categorization by Bengaly et al. [19]. These observed signs and animal death could be owing to the facts that highly virulent parasites tends to cause a severe damage as they multiply rapidly and reach a peak of parasitaemia within a short time after their introduction to a host.
The two strains of T. congolense behave contrariwise in cattle and mice. This could be due to variations of these hosts in terms of species, their body size, daily and metabolic activities and their immunological response. However, the observed high and low transmissibility and pathogenicity of T. congolenseresistant strain in mice and cattle respectively could also be attributed to pathogen defense against the hosts reactions and transmission cycle between hosts as this involved serial syringe passage of the stocks in mice prior to infection in cattle. Van Den Bossche et al. [5] suggested that susceptible host infected by highly virulent trypanosome strains will display a severe disease and leading to either treatment and/or death, leading therefore to a decrease in dispersion of highly virulent trypanosome as compared to its' less pathogenic competitor resulting in a relatively low tness. This complements the observed difference in pathogenicity of resistant strain in mice and cattle in this study.
Variation in the ability of mice and cattle in handling infection of these two trypanosome strains suggest variation in pathogenicity of the two strains attributed to a number of hosts-related factors. Neither parasite antigen switching nor developmental progression to transmission stages is driven by host, instead host contribute to the infection dynamics [27]. The level at which primary peak of parasitaemia is reached is host speci c [36]. Also, host's immunity to most variants cause drop in antigenic switching rate as infection progresses [37]. In consistence with our ndings, a number of studies correlated the ability of controlling anaemia with maintenance of low level of parasitaemia and thus signify trypanotolerance in cattle [30,38]. In contrast, trypanotolerance in mice associated with parasitaemia control as well as maintaining longer survival time [39]. As true this may be, results obtained in mice should cautiously be extrapolated in cattle, as may not be a true re ection of the same stock characteristics in cattle [40].

Conclusions And Recommendations
In conclusion, the pathogenicity of resistant and sensitive strain of T. congolense as re ected by different proxy parameters we monitored varied signi cantly across host species (mice and cattle). The resistant T. congolense strain was highly pathogenic in mice and less so in cattle. To the contrary, the drug sensitive T. congolense strain was highly pathogenic to cattle and less so to mice. As these variations observed, the present study therefore emphasize variations on the pathways by which different species and/or subspecies of trypanosomes act upon the host. Hence it is recommended that, with a large number of experimental animals; cattle in particular, various pathways can then be studied to represent the eld situation. Nevertheless, it would be worthwhile to take into consider ndings of this study when planning for control operations particularly with the use of diminazene aceturate and isometamidium chloride.

Declarations Acknowledgements
We deeply thank Prof. A.E. Kimambo for providing y-proof maintenance facility where the experimental steers were kept. We are also indebted to all the staff at the Animal Breeding Unit, Department of Microbiology, Parasitology and Biotechnology and Department of Animal, Aquaculture and Range Sciences who provided technical and logistical assistance at various stages of this research work. This research was funded by the government of Tanzania through the Commission for Science and Technology (COSTECH).

Funding
This study was supported through a PhD studentship granted to AFN by the Tanzania Commission for Science and Technology (COSTECH).

Data availability statement
Data generated in this work have been used to support the conclusions made in this study. However, the data will be made freely accessible to the readers upon request from the corresponding author.
Author's contributions AFN and LLM designed the study. AFN collected the data, analyzed them and interpreted the results. AFN and LLM wrote the manuscript. LLM conducted series of revisions on the manuscript and produced the nal version for submission. All authors read and approved nal version of the manuscript.

Ethics approval and consent to participate
The ethical clearance for conducting this study was obtained from the Research and Publication Committee of the Sokoine University of Agriculture, Morogoro, Tanzania.  This is a list of supplementary les associated with this preprint. Click to download.