11 Nov 2021: Chelbi I, Maghraoui K, Zhioua S, Cherni S, Labidi I, et al. (2021) Correction: Enhanced attraction of sand fly vectors of Leishmania infantum to dogs infected with zoonotic visceral leishmaniasis. PLOS Neglected Tropical Diseases 15(11): e0009947. https://doi.org/10.1371/journal.pntd.0009947 View correction
The sand fly Phlebotomus perniciosus is the main vector of Leishmania infantum, etiological agent of zoonotic visceral leishmaniasis in the Western Mediterranean basin. Dogs are the main reservoir host of this disease. The main objective of this study was to determine, under both laboratory and field conditions, if dogs infected with L. infantum, were more attractive to female P. perniciosus than uninfected dogs.
We carried out a series of host choice experiments and found that infected dogs were significantly more attractive to P. perniciosus than uninfected dogs in the laboratory as well as in the field. Significantly more P. perniciosus fed on infected dogs than on uninfected dogs. However, the fecundity of P. perniciosus fed on infected dogs was adversely impacted compared to uninfected dogs by lowering the number of laid eggs. Phlebotomus perfiliewi, the second most abundant sand fly species in the field site and a competent vector of L. infantum had similar trends of attractivity as P. perniciosus toward infected dogs under field conditions.
The results strongly suggest that L. infantum causes physiological changes in the reservoir host which lead to the host becoming more attractive to both male and female P. perniciosus. These changes are likely to improve the chance of successful transmission because of increased contact with infected hosts and therefore, infected dogs should be particularly targeted in the control of zoonotic visceral leishmaniasis in North Africa.
Zoonotic visceral leishmaniasis is a neglected tropical disease caused by the parasite Leishmania infantum. In the Western Mediterranean basin, Phlebotomus perniciosus and Phlebotomus perfiliewi are the main vectors of L. infantum, and dogs are the main reservoir host of the parasite. In Northern Africa, ZVL affect mostly children less than 5 years. Understanding the chemical ecology governing the relationships between the vector, the parasite and the reservoir host is of major epidemiological importance. Previous studies based on rodent models have shown that the infection with L. infantum enhanced the attractiveness towards Lutzomyia longipaplis, the main vector of ZVL in the Americas. In this study, we reported for the first time that infected dogs are highly attractive to both male and female P. perniciosus under laboratory and filed conditions compared to uninfected dogs. Similar patterns were observed with wild populations of P. perfiliewi. Our results provided strong evidence that the parasite manipulate the reservoir hosts to enhance its transmission success by the vector.
Citation: Chelbi I, Maghraoui K, Zhioua S, Cherni S, Labidi I, Satoskar A, et al. (2021) Enhanced attraction of sand fly vectors of Leishmania infantum to dogs infected with zoonotic visceral leishmaniasis. PLoS Negl Trop Dis 15(7): e0009647. https://doi.org/10.1371/journal.pntd.0009647
Editor: David Soeiro Barbosa, Universidade Federal de Minas Gerais, BRAZIL
Received: February 13, 2021; Accepted: July 13, 2021; Published: July 27, 2021
Copyright: © 2021 Chelbi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript.
Funding: EZ and AS received funding from the US National Institutes of Health -grant number R21AI130485, and from the Wellcome Trust-grant number 209336/Z/17/7. The funders had no role in study design, data collection or analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Zoonotic visceral leishmaniasis (ZVL) is a vector-borne zoonotic disease caused by the parasite Leishmania infantum, which is transmitted by the bite of female phlebotomine sand flies. ZVL can affect both humans and canines and is considered by the WHO to be one of the most important neglected tropical diseases. It affects about 0.5 million people per year [1,2] and is widespread in South and Central America, North Africa, Southern Europe, Middle and Far eastern countries. ZVL distribution is strongly correlated with poverty  and in Tunisia is a peri-domestic disease mostly endemic in rural areas affecting families of low social and economic status .
ZVL is a systemic disease that frequentlyresults in the death of infected individuals if untreated. No effective vaccine for human visceral leishmaniasis is available  and in Tunisia, the disease has an estimated incidence between 100 and 160 cases per 100,000 inhabitants  with a mortality rate of 6% . The incidence of ZVL is highest among children between 1 and 2 years of age . In a previous study, we showed that paediatric patients admitted 15 days after onset of symptoms, with bleeding, white cell counts below 4,000/mm3, and cytolysis at admission should be considered severe cases and subsequently, they are at high risk of mortality .
Domestic dogs are the main reservoir host for L. infantum in the Old World [8,9] and in the New World  and sand flies of the subgenus Larroussius, predominantly Phlebotomus perniciosus, are the main vectors of ZVL in Tunisia [9,11,12]. Other sand fly species of the subgenus Larroussius, mainly Phlebotomus perfiliewi, also play an important role in the transmission of ZVL throughout the Mediterranean basin including Tunisia [12–14]. In previous studies in Tunisia we showed that the prevalence of ZVL in dogs is an important parameter for determining transmission to humans [4,15]. Therefore, understanding the dynamics of transmission of the parasite between the canine reservoir host and the sand fly vector P. perniciosus is of major epidemiological importance.
Sand flies use odours, heat, CO2 of the host [16–18], and sex/aggregation pheromones emitted by male sand flies [19,20] to identify and orientate towards potential host animals for blood-meal and mate acquisition. While P. perniciosus can feed on a wide variety of vertebrate hosts to obtain blood meals , dogs remain the main reservoir host for L. infantum [4,9]. From an eco-epidemiological view, the parasite requires an overlap between the vector (P. perniciosus) and the reservoir host (dogs), which is a prerequisite for the emergence of a ZVL focus. To achieve this overlap, some parasites are known to manipulate the host animal by changing its odour or behaviour to improve their chances of transmission . It was shown that following infection with L. infantum, hamsters became significantly more attractive to females Lutzomyia longipalpis, vector of ZVL in South America [23,24].
We hypothesized that physiological changes in dogs parasitized by L. infantum change the dog’s odour making them more attractive to P. perniciosus and therefore enhancing the parasite’ s transmission success. To test this hypothesis, attractiveness of infected and uninfected dogs to P. perniciosus were assessed under both laboratory and field conditions.
Materials and methods
The maintenance of animals and the experimental procedures used in this research program followed the Animal Care and Use Protocol which is approved by the Institutional Animal Care and Use Committee of the Institut Pasteur de Tunis, Tunisia (2018/01/I/ES/IPT/V0). Infected dogs used in this research program were obtained from a previous study that was approved by the Institutional Animal Care and Use Committee of the Institut Pasteur de Tunis, Tunisia (IPT/UESV/27/2012). This work was performed under the Assurance of the US Office of Laboratory Animal Welfare [Assurance approval F-16-00170 (A5743-01)]. The Institut Pasteur de Tunis complies with the European Directive for the Protection of Vertebrate Animals used for experimental and other scientific purposes (2010/63/EU).
Assessing the attractiveness of female Phlebotomus perniciosus to infected and uninfected dogs under laboratory conditions
Sand flies used in this study were from a colony originated from Tunisia and maintained at the Laboratory of Vector Ecology in the Institut Pasteur de Tunis since 2003 [25,26]. Dogs used in the study were from the kennels of the Institut Pasteur de Tunis. We used 6 Beagle dogs that had been naturally infected with L. infantum as part of a different study to investigate the efficacy of a vaccine against canine ZVL. These dogs had been exposed to wild sand fly bites under natural conditions in a ZVL focus located at Borj Youssef in the governorate of Ariana (36°57’N, 10° 05’E) and were from the unvaccinated (control) group. Infection status of the infected dogs was confirmed by indirect immunofluorescent antibody test (IFAT) as described by Ben Slimane et al. (2014) . All Leishmania species isolated from field-collected female P. perniciosus in Tunisia were identified as L. infantum zymodeme MON-1 [11,21]. In Tunisia, the zymodeme MON-1 is responsible for the majority of human and canine cases [8,27]. The dogs were kept in the kennels after the exposure period and showed clinical signs 11 to 14 months after being exposed to wild sand fly bites. All symptomatic dogs showing specific signs of ZVL including lymphadenomegaly, hepatomegaly, splenomegaly, and progressive weight loss were classified as infected when introduced into our study. In addition, 6 uninfected Beagles from the kennel were used as controls.
The infectiousness the infected and uninfected dogs to sand flies was confirmed by xenodiagnosis . A minimum of 36% to 100% of the lab-colonised P. perniciosus that were fed on infected dogs developed L. infantum infections when dissected and examined under dissecting microscope as described by Chelbi and Zhioua (2019) , whereas none of the sand flies fed on uninfected dogs were found to be infected with parasites.
Infected dogs were housed individually in one part of the dog kennel where they received daily regular veterinary care. Uninfected dogs were housed in a separate part of the kennels. Protective measures were taken to avoid the infection of the uninfected dogs as described by Ben Slimane et al. (2014) . Each infected dog was paired by sex (6 males and 6 females) and age (varying from 3 to 4 years old) but not weight with an uninfected one. These pairs were then used in both the laboratory and field experiments.
To assess the preference of female P. perniciosus for infected or uninfected dogs, we carried out a simple choice experiment (Fig 1). The wire frames of three Barraud cages (cage A, B, and C) each measuring (40 x 40 x 40 cm) were welded to each other to create a frame that was 120 x 40 x 40 cm. Three netting cages were suspended in a line within the frame. The middle netting cage (B) was connected to the two cages on the left (A) and right (C) through openings (diameter x 10 cm) cut in the netting. One infected dog and its’ uninfected pair were anaesthetized by intramuscular (IM) injection of a mixture of 1.5 ml of ketamine (15 mg/Kg) (Merial, Lyon, France) and 0.02 ml/Kg of acepromazine maleate (Kela, N.V., Hoogstraten, Belgium). As ear skin is the best predictor of being infectious to vectors , the head of the infected and uninfected dogs were placed in the sections A and C respectively of the three connected-cages for 80 mins (Fig 1). A minimum of 200 females and 20 males uninfected laboratory-colonized P. perniciosus (F 29) were then released into the central part of the cage (B) and the females given the opportunity to choose either the infected or uninfected host in section A or C. Male sand flies were present to encourage female feeding [25,26] but we did not determine their preferences. Unfed sand flies were 5 to 7 days old and deprived of sugar for 24 hours prior to use in this experiment.
Access from the central part B of the apparatus to sections A and C was initially restricted by netting placed over the connecting openings. After the sand flies had been introduced into the central section, they were allowed to acclimatize for (20 mins) then the netting covering the connecting openings to both sections A and C were removed to allow the free movement of P. perniciosus toward either the infected or uninfected dog. After 20 minutes, the connecting openings between the central section B and side sections A and C were closed. The sand flies were then allowed to feed on the sedated dogs for 60 min in the dark at 27°C. After this time the number of engorged and unfed female P. perniciosus in section A and C were counted as well as the number of females in section B [considered as not responding (NR)] were counted. The experiment was replicated six times with six different pairs of infected and uninfected dogs. For each pair of dogs we carried out 5 replicates. A total of 30 replicates were done for this experiment. To control for any effect side bias, the positions of the infected and uninfected dogs were swapped every replicate. The work was performed in a bio-safety level laboratory 2.
We compared the fecundity of sand flies fed on infected and uninfected dogs. Engorged females were held individually in glass vials (5 cm high and 2.5 cm diameter) containing a small wet filter paper, with access to cotton wool soaked in a sucrose solution (30%). Vials were maintained at 27°C and 90% relative humidity in sealed polythene containers. Flies were examined daily for up to 10 days post-blood feeding; those in the infected groups that had died the previous day were dissected and examined for the presence of parasites in the gut. Only those flies in which parasites were observed (N = 75) were considered to be infected and used in the analysis. For each female sand fly, the number of eggs laid plus those retained in the ovaries after death were recorded. As this experiment is extremely time consuming, we compared the fecundity of a subset of sand flies fed on two pairs of infected and uninfected dogs. Thus with the first pair of dogs we examined 47 females engorged on the infected dog and 42 on the uninfected dog and for the 2nd pair of dogs 25 females engorged on the infected and 30 on the uninfected dog.
Assessing the attractiveness of Phlebotomus perniciosus to infected and uninfected dogs under field conditions
This study was performed in an endemic area for ZVL where P. perniciosus is the most abundant sand fly species followed by P. perfiliewi . The study took place during 9 consecutive nights in September 2019, a period corresponding to the main peak of activity of P. perniciosus and P. perfiliewi , at a rural dog kennel (36°58’N, 10°03’E) licensed by the Department of Agriculture and belonging to the Governorate of Ariana in northern Tunisia.
An experiment was carried out to determine the preference of primarily wild P. perniciosus and secondly of P. perfiliewi for either infected (symptomatic) or uninfected (asymptomatic) dogs. Three cages (100 × 90 × 100 cm) were placed on the ground, in a triangular pattern, equidistant (20 m) from each other (Fig 2). We used one pair of dogs that we had used in the previous laboratory experiments. In one cage, we placed an infected dog and in the 2nd we placed the uninfected paired dog. The 3rd cage was left empty as a negative control. Sand flies were collected at the cages using sticky traps made with, 13 white papers (20 cm x 20 cm; total area 1 m2) soaked in castor oil. The sticky traps were attached 1 m above the ground and evenly spaced along a cord around the top of the cage (Fig 2). The density is reported as the number of sand flies of each species per 1 m2 of sticky traps.
The animals were tested, between 18:00–06:00 HR and received water ad libitum during the night. The experiment was replicated 9 times and the cages were rotated between different positions to avoid positional bias. Sand flies collected using sticky traps were counted and identified to species level by using identification keys [32,33]. The atypical form of P. perniciosus females, often misidentified as P. longicuspis, were counted as P. perniciosus [34,35]. Other sand fly species collected on the traps were identified to species level using the same aforementioned keys.
To investigate sand fly preference for infected or uninfected dogs under laboratory conditions, we first compared numbers of female flies choosing side A or C by using a generalized linear model (GLM). A type II analysis of variance (ANOVA) was used to test for significant effects. The feeding success was analyzed using proportion data (females attracted / females feeding), because the input number of females in each cage was different. Feeding success was analysed with a GLM. Significant effects were tested with an ANOVA. Post-hoc analyses were performed with estimated marginal means (EMMEANS). Fecundity was analysed using a liner model and significant effects were determined with an ANOVA. Numbers of sand flies and eggs are given as mean ± SE. Similarly, proportions of fed sand flies (i.e. the number of females feeding divided by the number of females attracted) are given as mean ± SE.
To analyse the attractiveness of P. perniciosus to infected and uninfected dogs in the natural environment, a GLM with Poisson error was used to compare the number of sand flies between each treatment. Significant interactions and effects were tested using an ANOVA. Post-hoc analyses were performed using EMMEANS. The mean is the average number of specimens per variable in the different replicates. P-values less than 0.05 were considered to be significant and all analyses were performed using R v. 3.6.0.
Attractiveness of Phlebotomus perniciosus towards infected and uninfected dogs under laboratory conditions
The numbers of female P. perniciosus attracted towards infected dogs and uninfected dogs were 205.30 ± 16.94 and 80.57 ± 7.82 respectively and 18.80 ± 6.25 did not respond (NR) (Fig 3). The attractiveness of the dogs differed significantly according to their infection status (ANOVA: F = 66.709, DF = 2, P<0.001) and infected dogs attracted significantly more P. perniciosus females than uninfected dogs (EMMEANS: z = 5.989, P<0.001, Fig 3).
Y axis represents the number of sand flies collected from the cage baited with infected and from the cage with uninfected dog. X axis represents the infection status of the dogs. Boxes are limited by the minimal and maximal value. Error bars correspond to the standard error. Significance code: p≤0.001 ***.
The numbers of female P. perniciosus that fed on the infected and uninfected dogs were (186.16 ± 16.98), and (66.83 ± 7.58), respectively. The proportion of P. perniciosus females that fed on infected and uninfected dogs were 0.90 ± 0.02 and 0.81 ± 0.03 respectively (Fig 4). Feeding success was significantly affected by the infection status of dogs (ANOVA: F = 7.5714, DF = 1, P = 0.0079). The proportion of P. perniciosus females that fed on infected dogs was significantly higher than the proportion that fed on uninfected dogs (EMMEANS: z = 2.805, P = 0.005; Fig 4).
Y axis represents the proportion of engorged females collected from the cage baited with infected dog, and from the cage baited with uninfected dog. X axis represents the infection status of the dogs. Boxes are limited by the minimal and maximal value. Error bars correspond to the standard error. Significance code: p≤0.01 **.
The number of eggs laid by each female P. perniciosus fed on infected dogs and uninfected dogs were 36.46 ± 0.16, and 43.35 ± 0.15, respectively (Fig 5). Fecundity was highly affected by the infection status of dogs (ANOVA: F = 13.35, DF = 1, P<0.001). Females that had fed on infected dogs laid significantly fewer eggs than females that had fed on uninfected dogs (EMMEANS: t = -3.654, P<0.001).
Y axis represents the numbers of eggs laid by P. perniciosus fed on infected and uninfected dogs X axis represents the infection status of the dogs. Significance code: p≤0.001 ***.
The number of retained eggs in females that had fed on infected dogs was 2.35 ± 0.09 and in females fed on uninfected dogs was 1.27 ± 0.09 (Fig 6). Similarly, the infection status of the host affected the number of retained eggs (ANOVA: F = 73.605, DF = 1, P<0.001).
Attraction of sand flies to infected and uninfected dogs under field conditions
A total of 1,939 sand flies were collected during 27 trapping-nights. Phlebotomus perniciosus was the most abundant sand fly species (n = 1228, 63.3%) followed by P. perfiliewi (n = 662, 34.1%). The remaining sand fly species were Phlebotomus papatasi (n = 11, 0.56%), and Sergentomyia minuta (n = 38, 1.96%).
The numbers of P. perniciosus attracted to the infected dog, uninfected dog, and blank control were 20.44 ± 4.71, 46.78 ± 7.62, and 1.89 ± 0.63, respectively (Fig 7). Overall, there was a significant (ANOVA: F = 35.8120, DF = 2, P <0.001) difference in the numbers of P. perniciosus collected in the different cages. Infected dog attracted more P. perniciosus than uninfected dog (MCP: z = 3.666, P<0.001). Both infected and uninfected dogs attracted more P. perniciosus than empty cage (MCP: uninfected: z = 3.679, P<0.001; infected: z = 5.079, P<0.001).
Y axis represents the number of sand flies collected from the cage baited with infected dog, from the cage with uninfected dog, and from the un-baited cage. X axis represents the infection status of the dogs. Boxes are limited by the minimal and maximal value. Error bars correspond to the standard error. Significance code: p≤0.001 ***.
The number of females trapped was significantly higher (30.07 ± 6.43) than the number of males (16.00 ± 4.02) (MCP: z = 2.933, P = 0.00335) (Fig 8). The number of female P. perniciosus trapped in cages baited with the infected dog (59.44 ± 11.82) was significantly greater than the number of males trapped (34.11 ± 8.18) (EMMEANS: z = 2.933, P = 0.0393). Similarly, the number of female P. perniciosus trapped in the cage baited with the uninfected dog (29.11 ± 7.31) was significantly greater than the number of males trapped (11.78 ± 4.70) (EMMEANS: z = 2.933, P = 0.0393). The number of male P. perniciosus trapped in cages baited with the infected dog (34.11±8.18) was significantly greater than the mean number of males trapped in the cage baited with the uninfected dog (11.78 + 4.70) (EMMEANS: z = 3.666, P = 0.0034) (Fig 8). Similarly, the mean number of females P. perniciosus trapped in the cage baited with the infected dog (59.44 + 11.82) was significantly higher than the mean number of females trapped in cages baited with the uninfected dog (29.11 + 7.31) (EMMEANS: z = 3.666, P = 0.0034).
Y axis represents the number of sand flies collected from the cage baited with infected dog, from the cage with uninfected dog, and from the un-baited cage. X axis represents the infection status of the dogs. Boxes are limited by the minimal and maximal value. Error bars correspond to the standard error. Significance code: p≤0.01 **; p≤0.001 ***.
The numbers of P. perfiliewi attracted towards the infected dog, uninfected dog, and empty control cage were 27.22 ± 8.79, 9.39 ± 2.59 and 0.17 ± 0.09, respectively (Fig 9). Overall, the difference between numbers of P. perifiliewi collected in the cages with the infected dog, uninfected dog, and empty control cage was statistically significant (ANOVA: F = 94.715, DF = 2, P <0.001). Infected dog attracted more P. perfiliewi than uninfected dog (EMMEANS: z = 7.931, p<0.001). Both infected and uninfected dogs attracted more P. perfiliewi than empty control cage (EMMEANS: infected: z = 13.440, P<0.001; uninfected: z = 9.005, P<0.001).
Y axis represents the number of sand flies collected from the cage baited with infected dog, from the cage with uninfected dog, and from the un-baited cage. X axis represents the infection status of the dogs. Boxes are limited by the minimal and maximal value. Error bars correspond to the standard error. Significance code: p≤0.001 ***.
Overall, the numbers of male and female P. perfiliewi trapped were 4.59± 1.49 and 19.53±6.29, respectively. The mean number of females trapped was significantly higher than the number of males (EMMEANS: z = 10.52, P<0.001). The number of female P. perfiliewi trapped in the cage baited with infected dog (44.22 ± 15.56) was significantly greater than the number of males trapped (10.22 ± 3.79) (EMMEANS: z = 10.425, P<0.001). Similarly, the mean number (mean ± SE) of female P. perfiliewi trapped in the cage baited with the uninfected dog (15.44 ± 4.30) was significantly greater than the number of males trapped (3.33 ± 0.88) (EMMEANS: z = 7.344, P<0.001). The number of male P. perfiliewi trapped in the cage baited with the infected dog was significantly greater than the mean number of males trapped in the cage baited with the uninfected dog (EMMEANS: z = 3.990, P = 0.0013). Similarly, the mean number (mean ± SE) of female P. perfiliewi trapped in the cage baited with the infected dog was significantly higher than the mean number of females trapped in the cage baited with the uninfected dog (EMMEANS: z = 9.260, P<0.001).
In this study we investigated the effect of infection with L. infantum on the attractiveness of dogs in both the laboratory and the field in a natural focus of transmission. We tested the attractiveness of six pairs of dogs, matched for sex and age, one dog was infected (and symptomatic) and the other uninfected in both studies. Our results showed that dogs infected with L. infantum were highly attractive to both female and male P. perniciosus. The sand flies had a choice of orientating towards either of the potential hosts, but in all six laboratory replicates (pairs of dogs) female P. perniciosus were significantly more attracted to the infected than uninfected dogs. Similarly, we observed the same response pattern when one of the pair of dogs was exposed to the wild population of P. perniciosus.
The feeding success of P. perniciosus is significantly higher on infected dogs compared to uninfected ones. However, P. perniciosus that had fed on the uninfected dogs laid significantly more eggs than those that fed on dogs infected with L. infantum. Similar results were observed when P. langeroni and L. longipalpis were fed artificially on L. infantum infected blood [36,37]. While the longevity of L. longipalpis was reduced after artificial feeding on L. infantum infected blood, no significant impact on the fecundity was reported . All the aforementioned studies were obtained from artificially infected sand flies. Since we used a natural parasite-vector-host system, our results strongly suggest that L. infantum infection exerted an adverse impact on the fecundity of P. perniciosus. Resources for egg production may be diverted to limit the reduced longevity following infection with L. infantum. This hypothesis deserves further investigation.
Our study did not fully determine if the enhanced attractiveness of infected dogs that we observed was the result of host odour, thermal, visual or acoustic (or a combination of some or all of them) cues. However it is very likely that the sand flies are responding predominantly to odour cues. Several studies have suggested that dog odours are altered by L. infantum infection. Magalhaes et al. (2014)  identified 35 volatile organic components (VOCs) emitted by infected dogs that were either quantitatively or qualitatively different to uninfected dogs and could be recognised as bio-markers of infection. Similarly, a study using a VOC analyser (eNose) showed that dogs naturally infected with L. infantum in Brazil had a significantly different odour profile when compared to uninfected dogs .
Several other studies on non-natural infection systems that have also shown that odour of animals infected with L. infantum plays an important role in increasing the attractiveness of the infected animal. Nevatte et al. (2017)  showed that the attractiveness of golden hamsters (which do not naturally become infected with L. infantum) to L. longipalpis increased significantly after infection. The change in attractiveness was related to stage of infection and the change was more pronounced in some individuals than others .
Similar results have been obtained with other vector-parasite-host systems. Infection of the mouse with the parasite Plasmodium chabaudii, the etiological agent of rodent malaria, induces an increased attractiveness of Anopheles stephensi . A clear difference in the VOCs of mice infected with P. chabaudii compared to those of uninfected mice has been shown . Likewise, humans infected with P. falciparum are more attractive to Anopheles gambiae . However, there is evidence that the effect may not be universal e.g. there was no significant difference in the attractiveness of the sand fly Nyssomyia neivei toward BALB/c mice infected with Leishmania braziliensis and uninfected mice .
In order to confirm our laboratory results, we tested the attractiveness of dogs infected with L. infantum under natural conditions in a highly endemic focus for canine leishmaniasis. Our results showed that the number of P. perniciosus collected from the cage housing symptomatic L. infantum infected dog was significantly higher than the number of flies collected from the cage baited with uninfected dog. We observed the same pattern of responses from the wild population of P. perfiliewi to the same pair of dogs. This evidence strongly suggests that infected dogs are more attractive to both species of sand flies. The numbers of non L. infantum vectors species attracted to the dogs in the field experiments was too low to determine if there was a preference for infected rather than uninfected dogs. Our manipulation hypothesis would predict that there would be no difference in attraction of non-vector species to infected and uninfected dogs.
The enhanced attractiveness of P. perniciosus and P. perfiliewi to dogs infected with L. infantum under natural conditions is most likely due to the difference in the kairomones produced by the infected and uninfected dogs. A consequence of this manipulation could be enhanced transmission success of the parasite L. infantum to the vector. It would be interesting in the future to determine when the enhanced transmission success occurs in relation to the parasite life cycle in the host.
Our study showed that although both females and males P. perniciosus were attracted to uninfected dogs they were significantly more attracted to infected dogs in both the lab and field-based study. The response of the males seen here contrasts with the response of male L. longipalpis seen in other studies where there was no increased attraction of males to infected hamster . This study overcomes the limitations of the previously reported work, as it was partly carried out in the field with the presence of wild sand flies in an endemic ZVL focus and thus is the closest representation of natural transmission.
Based on blood meal analysis P. perniciosus is seen to be opportunistic feeding on whichever hosts are available [21,44–47]. No field studies on P. perniciosus have been carried out to investigate their relative attractiveness to different hosts. There is some evidence that L. longipalpis chooses hosts on the basis of their relative size rather than species , it is generally also considered to be an opportunistic feeder. The general zoophilic feeding behaviour of P. perniciosus and L. longipalpis based on host availability rather than on attractiveness to specific hosts should lead to a dilution of the parasite as it is spread to non-competent hosts, and this would subsequently exert a zooprophylactic effect on the transmission of L. infantum . Our work illustrates a potential zoopotentiation effect exerted by infected reservoir hosts that are highly attractive to sand fly vectors of L. infantum. Furthermore, it strongly suggests that infected dogs are the main reservoir host, as the different odour components and/or different concentrations that are released are significantly attractive to P. perniciosus compared to other environmental odours. This odour manipulation ensures successful transmission of the parasite to the vector. Thus, chemical ecology governs the L. infantum-dog-P. perniciosus relationship, and consequently, it has a direct impact on the transmission dynamic of ZVL.
While it has been shown that Leishmania infection may influence the quantity of blood ingested and the frequency of sand fly blood meals, thereby increasing the transmission rate of the parasite [38,50], it is not known whether parasites can affect host attractiveness to sand flies. Some parasites are known to manipulate their host animals by changing its physiology or behaviour to improve their chance of transmission . Our results strongly suggest that the parasite changes the physiology of the dog so that it becomes more attractive to female P. perniciosus, thus helping to ensuring its successful transmission.
Several studies have reported that symptomatic dogs, infected with L. infantum are highly infectious to their sand fly vectors compared to oligosymptomatic and asymptomatic dogs [51–58]. One explanation for this difference may be related to the relative attractiveness of symptomatic infected and uninfected dogs to their sand fly vectors. It will be interesting in due course to determine if infected asymptomatic and oligosympotomatic dogs are as attractive to the sand fly vectors as symptomatic dogs.
Greater attractiveness of infected dogs compared to uninfected dogs would have major epidemiological significance. In ZVL endemic areas of Tunisia, up to 50% of dogs are infected with L. infantum , therefore understanding the mechanisms which underpin the difference in attractiveness could help in the development of new approaches to reduce the infection rate of the vector, and subsequently to reduce the transmission of the parasite. It is also important to understand how the parasite may manipulate transmission and thus the effect on transmission models.
The authors would like to thank Patrick Schlupp for his valuable contribution to field work.
- 1. Desjeux P. Leishmaniasis: current situation and new prospectives. Comparative Immunology, Microbiology & Infectious Diseases. 2004; 27(6): 305–318. https://doi.org/10.1016/j.cimid.2004.03.004
- 2. Alvar J, Vélez ID, Bern C, Herrero M, Desjeux P, Cano J, et al. Leishmaniasis worldwide and global estimates of its incidence. PLoS One. 2012; 7 (5): e35671. pmid:22693548
- 3. Alvar J, Yactayo S, Bern C. Leishmaniasis and poverty. Trends in Parasitology. 2006; 22(12): 552–557. pmid:17023215
- 4. Zoghlami ZE, Chouihi E, Barhoumi W, Dachraoui K, Massoudi Ben Helel K, Habboul Z, et al. Interactions between canine and human visceral leishmaniases in a holoendemic focus of Central Tunisia. Acta Tropica. 2014; 139: 32–38. pmid:25004438
- 5. Handman E. Leishmaniasis: current status of vaccine development. Clinical Microbiology Reviews. 2001; 14(2): 229–243. pmid:11292637
- 6. Belhadj S, Djait-Baraket Z, Jemli B, Benosman A, Chaker E. Leishmanioses viscérales et cutanées du Nord: Etude rétroactive des cas diagnostiques à l’hôpital La Rabat de Tunis. Bulletin de la Société de Pathologie Exotique.1996; 89(4): 269–273. pmid:9053047
- 7. Ben Helel K, Ben Rejeb M, Habboul Z, Khattat N, Mejaouel H, Said-Latiri H,et al. Risk factors for mortality of children with zoonotic visceral leishmaniasis in Central Tunisia. PLoS One. 2017; e0189725. pmid:29287082
- 8. Aoun K, Diwani M F, Benikhlef R, Bouratbine A, Benhadjali S, Harrat Z, et al. Leishmania infantum MON-1: seul zymodème isolé chez les chiens leishmaniens en Tunisie. Bulletin de la Société de Pathologie Exotique. 2003; 96(2): 77–79. pmid:12836518
- 9. Ben Slimane T, Chouihi E, Ben Hadj Ahmed S, Chelbi I, Barhoumi W, et al. An investigation on vertical transmission of Leishmania infantum in experimentally infected dogs and assessment of offspring’s infectiousness potential by xenodiagnosis. Veterinary Parasitology. 2014; 206 (3–4): 282–286. pmid:25468025
- 10. Quinnell RJ, Courtenay O. Transmission, reservoir hosts and control of zoonotic visceral leishmaniasis. Parasitology. 2009; 136: 1915–1934. pmid:19835643
- 11. Ben Ismail R. Incrimination de Phlebotomus perniciosus comme vecteur de Leishmania infantum. Archives Institut Pasteur Tunis. 1993; 70(3): 91–110.
- 12. Barhoumi W, Fares W, Cherni S, Derbali M, Dachraoui K, Chelbi I, et al. Changes of sand fly populations and Leishmania infantum infection rates in an irrigated village located in arid Central Tunisia. International Journal of Environmental Research and Public Health. 2016; 13(3): 329. https://doi.org/10.3390/ijerph13030329.
- 13. Maroli M, Gramiccia M, Gradoni L. Natural infections of sand fly Phlebotomus perfiliewi with Leishmania infantum in a cutaneous leishmaniasis focus of the Abruzzi region, Italy. Transactions of the Royal Society of Tropical Medicine and Hygiene. 1987; 81(4): 596–598. pmid:3445341
- 14. Izri MA, Belazzoug S. Phlebotomus (Larrousius) perfiliewi naturally infected with deromtropic Leishmania infantum at Tenes, Algéria. Transactions of the Royal Society of Tropical Medicine and Hygiene. 1993; 87(4): 399. pmid:8249062
- 15. Kaabi B, Zhioua E. Modeling and comparative study of the spread of zoonotic visceral leishmaniasis from Northern to Central Tunisia. Acta Tropica. 2018; 178: 19–26. pmid:29050950
- 16. Dougherty HJ, Guerrin PM, Ward RD, Hamilton JGC. Behavioural and electrophysiological responses of the phlebotomine sandfly Lutzomyia longipalpis (Diptera: Psychodidae) when exposed to canid host odour kairomones. Physiological Entomology. 1999; 24(3): 251–262. https://doi.org/10.1046/j.1365-3032.1999.00139.x
- 17. Machado VE, da Rocha Silva FB, de Oliveira CM, Bueno AC, Francisco AA, Ortiz DGS, Goulart TM, Pinto MC. Ecologia química de flebotomíneos chpt 9 in: Atualidades em Medicina Tropical no Brasil: Vetores eds: Jader de Oliveira et al. 2020; https://doi.org/10.35170/ss.ed.9786586283129
- 18. Nigam Y, Ward RD. The effect of male sandfly pheromone and host factors as attractants for females Lutzomyia longipalpis (Diptera: Psychodidae). Physiological Entomology. 1991; 16(3): 305–312. https://doi.org/10.1111/j.1365-3032.1991.tb00569.x
- 19. Bray DP, Hamilton JGC. Host odor synergizes attraction of virgin female Lutzomyia longipalpis (Diptera: Psychodidae). Journal of Medical Entomology. 2007;44(5): 779–787. pmid:17915508
- 20. Spiegel CN, dos Santos Dias DB, Araki AS, Hamilton JGC, Brazil RP, Jones TM. The Lutzomyia longipalpis complex: a brief natural history of aggregation-sex pheromone communication. Parasites &Vectors. 2016; 9:580. pmid:27842601
- 21. Chargui N, Slama D, Haouas N, Rmadi L, Babba H. Transmission cycle analysis in a Leishmania infantum focus: Infection rates and blood meal origins in sand flies (Diptera: Psychodidae). Journal of Vector Ecology. 2018; 43(2): 321–327. pmid:30408299
- 22. Poulin R. “Adaptive” changes in the behaviour of parasitized animals: A critical review. International Journal for Parasitology. 1995; 25(12): 1371–1383. pmid:8719948
- 23. O’Shea B, Rebollar-Tellez E, Ward RD, Hamilton JGC, el Naiem D, Polwart A. Enhanced sandfly attraction to Leishmania-infected hosts. Transactions of the Royal Society of Tropical Medicine and Hygiene2002; 96(2): 117–118. pmid:12055795
- 24. Nevatte TM, Ward RD, Sedda L, Hamilton JGC. After infection with Leismania infantum, golden hamsters (Mesocricetus auratus) become more attractive to female sandflies (Lutzomyia longipalpis). Scientific Reports. 2017; 7,6104, pmid:28733676
- 25. Maroli M, Fiorentino S, Guandalini E. Biology of a laboratory colony of Phlebotomus perniciosus (Diptera: Psychodidae). Journal of Medical Entomology. 1987; 24: 547–551. pmid:3669026
- 26. Chelbi I, Zhioua E. Biology of Phlebotomus papatasi Scopoli (Diptera:Psychodidae) in the laboratory. Journal of Medical Entomology. 2007; 44(4): 597–600. pmid:17695013
- 27. Bouratbine A, Aoun K, Gharbi M, Haouas N, Zaroui J, Harrat Z, et al. Données épidémiologiques, cliniques et parasitologique sur la leishmaniose générale canine en Tunisie. Bulletin de la Société de Pathologie Exotique. 2005; 98: 359–362. pmid:16425713
- 28. Chelbi I, Zhioua E. Establishment, Maintenance of Phlebotomus spp. in the Laboratory, and Infection with Leishmania spp. Methods in Molecular Biology, Leishmania. 2019; 351–368.
- 29. Courtenay O, Carson C, Calvo-Bado L, Garcez LM, Quinnell RJ. Heterogeneities in Leishmania infantum infection: Using skin parasite burdens to identify highly infectious dogs. PLoS Neglected Tropical Diseases. 2014; 8,e2583. pmid:24416460
- 30. Zhioua E, Kaabi B, Chelbi I. Entomological investigations following the spread of visceral leishmaniasis in Tunisia. Journal of Vector Ecology. 2007; 32(2): 371–374. pmid:18260530
- 31. Bichaud L, Dachraoui K, Piorkowski G, Chelbi I, Moureau G, Cherni S, et al. Isolation of Toscana virus from sand flies, Tunisia. Emerging Infectious Diseases. 2013; 19(2): 322–324. pmid:23460990
- 32. Croset H, Rioux JA, Master M, Bayar N. Les phlébotomes de la Tunisie (Diptera, Phlebotominae). Mise au point systématique, chorologique et éthologique. Annales de Parasitologie Humaine and Comparée. 1978; 53(6): 711–749. https://doi.org/10.1051/parasite/1978536711
- 33. Léger N, Pesson B, Madulo-Leblond G, Abonnenc E. Sur la différentiation des femelles du sous-genre Larroussius Nitzulescu, 1931 (Diptera: Phlebotominae) de la région méditerranéenne. Annales de Parasitologie Humaine and Comparée. 1983; 58(6): 611–623. https://doi.org/10.1051/parasite/1983586611
- 34. Pesson B, Ready JS, Benabdennbi I, Martín-Sánchez J, Esseghir S, Cadi-Soussi M, Morillas-Marquez F, Ready PD. Sand flies of the Phlebotomus perniciosus complex: mitochondrial introgression and a new sibling species of Phlebotomus longicuspis in the Moroccan Rif. Medical and Veterinary Entomology. 2004; 18(1): 25–37. pmid:15009443
- 35. Boussaa S, Boumezzough A, Remy PE, Glasser N, Pesson B. Morphological and isoenzymatic differentitation of Phlebotomus pernicious and Phlebotomus longicuspis (Diptera: Psychodidae) in Southern Morocco. Acta Tropica. 2008; 106(3): 184–189. pmid:18456222
- 36. El Sawaf BM, El Sattar SA, Shehata MG, Lane RP, Morsy TA. Reduce longevity and fecundity in Leishmania-infected sandflies. American Journal of Tropical Medicine and Hygiene. 1994; 51: 767–770.
- 37. Agrela IF, Feliciangeli MD. Effect of Leishmania spp infection on the survival, life expectancy, fecundity, and fertility of Lutzomyia longipalpis s.l. and Lutzomyia pseudolongipalpis. Memorias Instituto Oswaldo Cruz. 2015; 110: 611–617.
- 38. Rogers ME, Bates PA. Leishmania manipulation of sand fly feeding behavior results in enhanced transmission. PLoS Pathogens. 2007; 3(6):e91. pmid:17604451
- 39. Magalhaes-Junior JT, Mesquita PRR, dos Santos Olivera WF, Santos Olivera F, Franke CR, de Medeiros Rodrigues F, et al. Identification of biomarkers in the hair of dogs: new diagnostic possibilities in the study and control of visceral leishmaniasis. Analytical and Bioanalytical Chemistry. 2014; 406: 6691–6700. pmid:25171830
- 40. Staniek ME, Sedda L, Gibson TD, de Souza CF, Costa EM, Dillon RJ, et al. eNose analysis of volatile chemicals from dogs naturally infected with Leishmania infantum in Brazil. PLoS Neglected Tropical Diseases. 2019; 13, e0007599. pmid:31386662
- 41. De Moraes CM, Stanczyk NM, Betz HS, Pulido H, Sim DG, Read AF, et al. Malaria-induced changes in host odours enhance mosquito attraction. Proceedings of the National Academy of Sciences USA. 2014; 111(30): 11079–11084. pmid:24982164
- 42. Lacroix R, Mukaban WR, Gouagna LC, Koella JC. Malaria infection increases attractiveness of humans to mosquitoes. PLoS Biology. 2005; 3(9): e298. https://dor.org/ 10.1371/journal.pbio.003029843. pmid:16076240
- 43. Rocha Silva FB, Miguel DC, Machado VE, Oliveira WHC, Goulart TM, Tosta CD, et al. Influence of Leishmania (Viannia) braziliensis infection on the attractiveness of BALB/c mice to Nyssomyia neivai (Diptera:Psychodidae). PLoS One. 2019; 14(4), e0214574. pmid:30934013
- 44. Bongiorno G, Habluetzel A, Khoury C, Maroli M. Host preferences of phlebotomine sand flies at a hypoendemic focus of canine leishmaniasis in central Italy. Acta Tropica. 2008; 88(2): 109–116. https://doi.org/10.1016/s0001-706x(03)00190-6
- 45. Maia C, Dionísio L, Afonso MO, Neto L, Cristóvão JM, Campino L. Leishmania infection and host-blood feeding preferences of phlebotomine sandflies and canine leishmaniasis in an endemic European area, the Algarve Region in Portugal. Memórias do Instituto Oswaldo Cruz. 2013; 108(4): 481–487. pmid:23827997
- 46. Bravo-Barriga D, Parreira R, Maia C, Afonso MO, Blanco-Ciudad J, Serrano FJ, et al. Detection of Leishmania DNA and blood meal sources in phlebotomine sand flies (Diptera: Psychodidae) in western of Spain: update on distribution and risk factors associated. Acta Tropica. 2016; 164: 414–424. pmid:27720626
- 47. Bennai K, Tahir D, Lafri I, Bendjaballah-Laliam A, Bitam I, Parola P. Molecular detection of Leishmania infantum DNA and host blood meal identification in Phlebotomus in a hypoendemic focus of human leishmaniasis in northern Algeria. PLoS Neglected Tropical Diseases. 2018; e0006513. pmid:29958283
- 48. Quinnell RJ, Dye C, Shaw JJ. Host preferences of the phlebotomine sandfly Lutzomyia longipalpis in Amazonian Brazil. Medical and Veterinary Entomology. 1992;6(3),195–200. pmid:1421498
- 49. Alexander B, Lopes de Carvalho R, McCallum H, Pereira MH. Role of the domestic chicken (Gallus gallus) in the epidemiology of urban visceral leishmaniasis in Brazil. Emerging and Infectious Diseases. 2002; 8 (12):1480–1485. pmid:12498667
- 50. Rogers ME. The role of Leishmania proteophosphoglycans in sand fly transmission and infection of the mammalian host. Front in Microbiology. 2012; 3: 223, pmid:22754550
- 51. Gradoni L, Maroli M, Gramicia M, Mancianti F. Leishmania infantum infection rates in Phlebotomus perniciosus fed on naturally infected dogs under antimonial treatment. Medical and Veterinary Entomology. 1987; 1(4): 339–342. pmid:2979550
- 52. Molina R, Amela C, Sand-Andres M, Gonzales F, Castillo JA, Lucientes J, et al. Infectivity of dogs naturally infected with Leishmania infantum to colonized Phlebotomus perniciosus. Transactions of the Royale Society of the Tropical Medicine and Hygiene. 1994; 88(4): 491–493. pmid:7570854
- 53. Guarga JL, Lucientes J, Peribanez MA, Molina R, Garcia MJ, Castillo JA. Experimental infection of Phlebotomus perniciosus and determination of the natural infection rates of Leishmania infantum in dogs. Acta Tropica. 2000; 77(2): 203–207. pmid:11080511
- 54. Guarga JL, Moreno J, Lucientes J, Gracia MJ, Peribanez MA, Alvar J, et al. Canine leishmaniasis transmission: high infectivity amongst naturally infected dogs to sand flies is associated with lower proportion of T helper cells. Research in Veterinary Sciences. 2000; 69(3): 249–253. https://doi.org/10.1053/rvsc.2000.0419
- 55. Vexenat JA, Fonseca de Castro JA, Cavalcante R, Tabares JP, da Silva MRB, Furtado Campos JH, et al. Visceral leishmaniasis in Teresina, State of Piaui, Brazil: Preliminary observations of the detection and transmissibility of canine and sandfly collections. Memórias do Instituto Oswaldo Cruz. 1994; 82(2): 131–135. https://doi.org/10.1590/S0074-02761994000200001
- 56. Tarvi BL, Tabares CJ, Cadena H, Ferro C, Osorio Y. Canine visceral leishmaniasis in Colombia: relationship between clinical and parasitological status and infectivity for sand flies. American Journal of Tropical Medicine and Hygiene. 2001; 64(3–4): 119–124. https://doi.org/ 10.4269/ajtmh.2001.64.119
- 57. Da Costa-val AP, Cavalcanti RR, De Figueiredo Contijo N, Michalick MSM, Alexander B, Williams P, et al. Canine visceral leishmaniasis: relationship between clinical status, humoral immune response, haematology and Lutzomyia (Lutzomyia) longipalpis infectivity. Veterinary Journal. 2007; 174(3): 636–643. https://doi.org/10.1016/j.tvjl.2006.11.006.
- 58. Michalsky EM, Rocha MF, da Rocha Lima ACVM, França-Silva JC, Pires MQ, Olivera FS, et al. Infectivity of seropositive dogs, showing different clinical forms of leishmaniasis, to Lutzomyia longipalpis phlebotomine sand flies. Veterinary Parasitology. 2007; 147(1–2): 67–76. pmid:17449184