- Email: [email protected]
Seasonal dynamics of ticks (Acari: Ixodidae) on horses in the state of São Paulo, Brazil
Veterinary Parasitology 105 (2002) 65–77
Seasonal dynamics of ticks (Acari: Ixodidae) on horses in the state of São Paulo, Brazil Marcelo B. Labruna a,∗ , Nobuko Kasai a , Fernando Ferreira a , João L.H. Faccini b , Solange M. Gennari a a
Faculdade de Medicina Veterinária e Zootecnia, Departamento de Medicina Veterinária Preventiva e Saúde Animal, Universidade de São Paulo, Av. Prof. Orlando Marques de Paiva, 87 Cidade Universitária, São Paulo, SP 05508-000, Brazil b Departamento de Parasitologia Animal, Instituto de Veterinária, Universidade Federal Rural do Rio de Janeiro, Km 7, BR 465, Seropédica, RJ 23890-000, Brazil Received 27 July 2001; received in revised form 20 November 2001; accepted 26 November 2001
Abstract Natural tick infestations were assessed every 14 days on horses over a 2-year period. Amblyomma cajennense adult ticks were counted individually, without detachment from the horses. Larvae and nymphs of A. cajennense were collected using a rubber scraper that scratched engorged immature ticks from the host. Adult females of Anocentor nitens larger than 4 mm length were counted on the horses. Blood samples were also obtained from the horses every 14 days and macroclimatic data were obtained for the study period. Infestations of A. cajennense demonstrated distinct peaks of activity for each of the three parasitic stages over each 12-month period, showing a 1-year generation pattern. Larvae predominated from April to July and nymphs from June to October. Adults predominated from October to March with a greater number of adult males than females. Although other studies on seasonal dynamics in the states of Rio de Janeiro and Minas Gerais were performed with the free-living stages of A. cajennense on pastures, the present study in the state of São Paulo, performed with the parasitic stages of A. cajennense on horses, showed similar results to those observed in other states. Infestations by A. nitens demonstrated distinct peaks of activity of adult females (>4 mm), suggesting different tick generations during the year. Infestation with A. nitens was much higher in the first year than the second year which may have been related to horse nutritional status and stocking rate. Although several climatic variables showed statistical significant correlation (r) with tick counts, the determination coefficients (R2 ) were always lower than 0.40, suggesting that any single significant variable (i.e. mean temperature) would not explain the tick distribution pattern over the year. The highest peaks of A. nitens females (>4 mm) were significantly associated with decrease in horse packed cell volumes (R 2 = 0.603). The ears and
∗ Corresponding author. Tel.: +55-11-3818-7703; fax: +55-11-3818-7028. E-mail address: [email protected] (M.B. Labruna).
0304-4017/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 4 0 1 7 ( 0 1 ) 0 0 6 4 9 - 5
66
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
the perineum, tail and groin region accounted for around 70% of all A. nitens females counted on the horses. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Anocentor nitens; Amblyomma cajennense; Seasonal dynamics; Horse; Brazil
1. Introduction Several studies have reported that three tick species infest horses in Brazil: Anocentor nitens, Amblyomma cajennense and less commonly, Boophilus microplus (Moreno, 1984; Falce, 1986; Heuchert et al., 1999; Labruna et al., 2001). Horses act as primary hosts for A. nitens and A. cajennense, as the establishment of populations of both species in areas inhabited by horses is not dependent on the presence of other host species (Labruna et al., 2001). Conversely, horses are secondary hosts for B. microplus, as the presence of this tick is dependent on the presence of its primary host cattle, grazing with horses in the same area (Moreno, 1984; Heuchert et al., 1999; Labruna et al., 2001). A. nitens is a one-host tick and was shown to be the most common tick species on horses in the state of São Paulo (Labruna et al., 2001). It attacks the host, particularly the region of the ear, nasal diverticulum, mane and perineum (Borges et al., 2000). Infestations can cause severe lesions in the ears that predispose the host to bacterial infections and screw-worm infestations. A. nitens is the natural vector of Babesia caballi in Latin America (Roby and Anthony, 1963). A. cajennense, a three-host tick, is also implicated as a serious ectoparasite of horses and is the principal tick infesting humans in central and southern Brazil (Aragão, 1936; Labruna et al., 2001). A. cajennense is the main vector of Rickettsia rickettsii, the agent of human Brazilian spotted fever (Dias and Martins, 1939). It was also proven to be a competent vector of the Venezuelan equine encephalomyelitis virus under laboratory conditions (Linthicum et al., 1991). In a 1-year period of field sampling of unfed larvae, nymphs and adults of A. cajennense in horse pastures in the state of Rio de Janeiro (Serra-Freire, 1982), it was suggested that this tick completed one generation a year. This proposition was later confirmed in the studies of Souza (1990) in the state of Rio de Janeiro and Oliveira et al. (2000) in the state of Minas Gerais. Studies regarding the seasonal dynamics of A. nitens are limited. Souza (1990) and Borges et al. (2000) reported 3–4 peaks of host seeking larvae on the pasture or adult females on the horses during the year, throughout a 2-year study in the states of Rio de Janeiro and Minas Gerais, respectively. They suggested that A. nitens completed 3–4 generations per year in the studied areas. This paper reports the seasonal dynamics of tick infestations on horses over a 2-year period in the state of São Paulo. 2. Materials and methods 2.1. Study site and horses This study was conducted at the experimental ranch of the Veterinary Faculty of the University of São Paulo, located at Pirassununga County at 631 m of altitude (21◦ 59 24 S,
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
67
47◦ 25 12 W), state of São Paulo, Brazil. On September 5th, 1997, 10 crossbred horses, 3–7 years of age (mean: 4.5 years old), were kept in a 4 ha pasture that was naturally infested with A. cajennense and B. microplus, based on previous examination of cattle that occupied the pasture and were removed 4 weeks before the introduction of the study horses. The horses came from a commercial farm located at Pirassununga and were naturally infested with A. nitens at the time they were placed on the pasture. Pasture consisted of forage grasses (Paspalum notatum, Brachiaria decumbens) and also several bushes and shrub species (Acacia bonariensis, Ageratum conyzoides, Senna occidentalis and Sida spp). 2.2. Tick counts Evaluation of natural tick infestations on the horses started on October 15th, 1997 and ended on September 29th, 1999. A. nitens tick counts were continued for two additional months, until November 24th, 1999. Horses were never treated with acaricides during the experiment. On October 14th, 1998, five horses were endangered by high tick burdens and low packed cell volume (PCV) levels. These five horses were removed from the pasture. Therefore, the pasture was managed with 2.5 horses/ha in the first year and 1.25 horse/ha in the second year. During the experiment tick burdens on the horses were evaluated every 14 days, always on Wednesday mornings, totaling 52 tick count observations during the study for A. cajennense and 56 tick count observations for A. nitens. Counts of A. cajennense were performed according to the methodology described and tested by Oliveira (1998) for the three parasitic stages of A. cajennense on horses. Ticks were evaluated on the entire left side of the horse body and the total number of ticks was doubled to estimate the total number of ticks on each animal. Adult males and females of A. cajennense were counted individually, without detaching the ticks. Larvae and nymphs of A. cajennense were checked using a rubber scraper which scratched engorged immature ticks from the host body. This procedure removed only totally engorged immature ticks (Oliveira, 1998). Prior to counting the number of engorged immature ticks, they were maintained in the laboratory at 27 ◦ C for 7 days to allow engorged larvae of A. nitens to molt to nymphs. During the first 8 months of the experiment, all engorged larvae that died before molting were identified to generic level (Clifford and Anastos, 1960). Thus, larvae identified in the genus Amblyomma were considered to be A. cajennense, whereas larvae identified in the genus Dermacentor or Anocentor were considered to be A. nitens. A total of 1248 dead larvae were examined during the 8 months that this procedure was conducted. As only eight (0.6%) larvae belonged to the genus Anocentor, this procedure was discontinued and all engorged dead larvae were counted as A. cajennense. Although the differentiation of engorged nymphs of both species was easily performed by macroscopic visualization, the 7 day-period that the ticks were maintained at 27 ◦ C was also sufficient for engorged nymphs of A. nitens to molt to adults. Evaluation of A. nitens on the horses, was by the method described by Borges et al. (2000). Female ticks larger than 4 mm length were counted on the right side of each horse. For this purpose, the entire right side of the horse was divided into five areas which were designated as: (1) ear; (2) mane; (3) face and neck; (4) perineum, tail and groin; (5) thorax, ventrum and legs. Ticks present in the nasal diverticulum were not counted due to the difficulty of
68
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
assessing this anatomic area. The number of ticks in each of the five areas was doubled and summed to estimate the total number of ticks on each animal. Blood samples were obtained through the jugular vein into individual EDTA-coated glass tubes every 14 days for determination of the PCV values by the microhematocrit method (Archer, 1977). During the study period all horses were orally drenched with Oxibendazol (Equitac, Pfizer) every 2 or 3 months for control of helminth parasitism. 2.3. Environmental data Macroclimatic data for the study site were obtained from the meteorological station of the experimental farm, situated approximately 150 m from the pasture where the horses were maintained during the experiment. Daily measurements of maximum, minimum and median temperatures, total rainfall and median and minimum relative humidities (RHs) were obtained for the entire study period. Data on temperature and RH are presented as the mean of the previous 14 days to each tick count and data on rainfall as the sum of the previous 14 days for each tick count. Daylight length (photoperiod) for the study site over the study period was obtained from a Geo Position System electronic device (Carmin GPS 12 XL, Olathe, KS, USA). 2.4. Data analysis Scatter plots between the mean tick counts and climatic variables were performed to verify if there were any linear associations between the variables. Pearson correlation coefficients (r) were calculated for each of the climatic, PCVs and mean tick count values for the entire study period. All analyses were performed using SPSS for Windows (1999).
3. Results Counts of A. cajennense on horses demonstrated distinct peaks of activity for each of the three parasitic stages over each 12-month period (Fig. 1). Larval peaks were observed from April to July in both years, when 250,850 engorged larvae (98.2% of the total larvae) were counted. The highest larval peaks occurred in May when mean tick counts were 5645±2151 larvae per horse in the first year and 5798 ± 3802 per horse in the second year (Fig. 1). From October to February of both years, no A. cajennense larvae were found on horses. Nymphal peaks occurred from June to October in both years when 34,794 engorged nymphs (99.5% of the total nymphs) were counted. The highest nymphal peaks occurred in August and September when mean tick counts were 406.4 ± 648 nymphs per host in the first year and 850.4 ± 1144 in the second year. No engorged nymph was found on horses from January to April in the first year and only 12 engorged nymphs were counted during this same period in the second year. Peaks of A. cajennense adult ticks occurred from October to March when 54,556 ticks (83.7% of the total adults) were counted and the mean tick counts were always higher than 100 adult ticks per horse in the first year and higher than 200 in the second year. The highest peak of A. cajennense adult ticks on horses occurred in February 1999 when a mean of 1053 ± 413.5 adult ticks per horse was found. Despite showing distinct peaks from
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
69
Fig. 1. Seasonal dynamics of A. cajennense on horses pastured on a ranch in Pirassununga, São Paulo. Infestations by adults (A), larvae (B) and nymphs (C) were evaluated on the horses every 14 days from October 1997 to September 1999. In graph A, the plotted line represents the mean number of adults on the horses and the bars represent the mean ratio of the number of adult males over females on the horses.
70
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
Fig. 2. Climatic data at Pirassununga County from September 1997 to November 1999.
September to March in both years, the A. cajennense adult stage was found on horses at all count dates of the study. However, adult tick numbers were very low and at many times near zero in counts performed at times other than September–March (Fig. 1). Comparisons between climatic data obtained for the study period (Fig. 2) and tick count data presented in Fig. 1 showed that the periods of highest activities of A. cajennense larvae and nymphs were the autumn and winter months, when there were shorter daylight periods and lower mean temperatures, RH and rainfall. On the other hand, the periods of highest activity of the A. cajennense adult stage occurred during the spring and summer months when there were longer daylight periods and higher mean temperatures, RH and rainfall. The correlation coefficients between the mean tick counts and the climatic data for the study period are shown in Table 1. Engorged larvae counts showed significant negative correlation (P < 0.05) with the mean values of maximum, minimum and mean temperatures, and daylight hours. Similarly, engorged nymph counts showed significant
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
71
Table 1 Results of statistical correlation analysis between the climatic variables and tick count values on horses over a 24-month period in Pirassununga, São Paulo (data are presented as the Pearson correlation coefficient (r) followed by the P value in parentheses) Climatic variables
Maximum temperature Minimum temperature Mean temperature Mean RH Minimum RH Total rainfall Photoperiod (light hours)
A. cajennense Larva
Nymph
Adult
−0.52 (<0.001) −0.55 (<0.001) −0.50 (<0.001) 0.04 (0.741) −0.16 (0.253) −0.25 (0.070) −0.56 (<0.001)
−0.22 (0.113) −0.38 (0.005) −0.51 (<0.001) −0.60 (<0.001) −0.62 (<0.001) −0.35 (0.100) −0.39 (0.004)
0.46 (0.001) 0.51 (<0.001) 0.56 (<0.001) 0.26 (0.054) 0.31 (0.022) 0.47 (<0.001) 0.55 (<0.001)
A. nitens (females >4 mm) −0.44 (0.001) −0.39 (0.003) −0.33 (0.011) 0.14 (0.290) 0.12 (0.928) −0.12 (0.456) −0.37 (0.004)
negative correlation (P < 0.05) with the mean values of minimum and mean temperatures, minimum and mean RH, and daylight hours. On the other hand, the adult counts showed significant positive correlation (P < 0.05) with the mean values of maximum, minimum and mean temperatures, minimum RH, total rainfall and daylight hours. The scatter plots indicated that none of the climatic variables presented a linear relationship with the A. cajennense counts, thus making unnecessary the application of the linear multivariable model. Although there were statistically significant correlations between the climatic data and the mean values of A. cajennense larvae, nymphs and adults on horses, the determination coefficients (R2 ) were always low (<0.40). In this case, R2 indicates the proportion of the tick count variance that is attributable to variation in the fitted mean of tick count values, as a function of the climatic data. The results suggested that any single significant variable alone (i.e. mean temperature) would not explain the tick distribution pattern over the year. Infestations by adult A. cajennense on the horses were characterized by a higher number of male than female ticks during all of the study period (Fig. 1A). During the period of highest adult activity (October–March) the proportion of males over females was higher in the first half of this period (October–December) than the second half (January–March). In the first year, this proportion was approximately four males for each female [4:1 (M:F)] from October to December 1997, dropping to about 2:1 (M:F) in the period from January to March 1998. Similarly, in the second year this proportion was about 2.5:1 (M:F) from October to December 1998 dropping to about 1.4:1 (M:F) during January–March 1999. The proportions of males to females during the rest of the year did not show a definitive pattern. These periods were characterized by the highest activity of immature ticks and few adult ticks were present on horses, some times only male ticks were observed (data not shown). The parasitism by A. nitens engorged females >4 mm was first recorded on the horses by the end of November 1997, 2 months after tick counts began (Fig. 3). For this reason, the A. nitens counts were continued until November 1999. The infestation pattern by A. nitens was characterized by different peaks over the year. The first peak of infestation occurred during December 1997, and after that, four other peaks through November 1998. In the first year, the highest peaks of A. nitens infestation occurred in April 1998 (mean: 273.3 ± 179.2 ticks per horse) and July 1998 (291.4 ± 196.4). A third peak of infestation occurred in October 1998 (219.0 ± 178.4). From November 1998 to November 1999, when the stocking rate
72
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
Fig. 3. Mean number of A. nitens females >4 mm in length (•) and mean PCV (䊏) evaluated on horses every 14 days from October 1997 to November 1999 at Pirassununga, São Paulo.
of the pasture had dropped to half of the first year, A. nitens counts were much lower as all counts showed mean values lower than 100 ticks per horse. Therefore, similar to the first year, the highest peaks of infestation in the second year occurred during the first semester followed by a drastic reduction of infestation in August 1999 and new higher peaks of infestation during September–October 1999. Although there were statistically significant negative correlations between the mean, maximum and minimum temperatures and photoperiod data with mean values of A. nitens on horses, the determination coefficients (R2 ) were also always low (<0.40). As the scatter plots did not indicate any linear relationship between the climatic variables presented in Fig. 2 and A. nitens counts, the application of a linear multivariable model was again unnecessary. The mean values of PCV over the study showed significant negative statistical correlation with A. nitens count values (P < 0.001), showing a R 2 = −0.603. This coefficient indicated that the A. nitens infestation explained 60.3% of the PCV variation observed on the horses. In fact, it can be seen in Fig. 3 that the higher A. nitens peaks coincided in time with more pronounced decreases of PCV mean values. The lowest mean PCV value (23.4±5.0%) was observed on July 1998 when the highest mean A. nitens count (291.4 ± 196.4) was also observed. The predilection sites for A. nitens infestation on the horse were the ears and the perineum, tail and groin region. From a total of 13,730 adult females >4 mm counted on the right side of the horses, 5086 (37.0%) were in the internal face of the right ear and 4414 (32.2%) in the perineum, tail and groin area. Mane and head accounted for 14.6 and 12.0% of the total ticks, respectively. Only 4.2% of the ticks were counted on thorax, ventrum and legs. During the first six tick counts of the study, from October to December 1997, an infestation of at least one horse with B. microplus was noted at each tick count. B. microplus infestation
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
73
was low with no more than six B. microplus adult ticks on each horse. After December 1997, no B. microplus ticks were observed on the horses for the remainder of the study.
4. Discussion Oliveira (1998) determined that infestations by the three parasitic stages of A. cajennense on horses showed a marked symmetry, making it possible to obtain an adequate estimate of population by counting only one side of the host. In addition, Oliveira (1998) determined that there was no significant difference between counting ticks for every 7 or 14 days. The A. cajennense population of São Paulo state demonstrated a 1-year generation pattern. Although other studies on seasonal dynamics in the states of Rio de Janeiro (Serra-Freire, 1982; Souza, 1990) and Minas Gerais (Oliveira et al., 2000) were performed with the free-living stages of A. cajennense on the pastures, the present study was concentrated on the parasitic stages of A. cajennense on horses and provided similar results to those observed in other states. In both studies, immature ticks predominated in the autumn/winter months and the adult stage predominated in the spring/summer months. Photoperiod and climatic variables such as temperature, RH and rainfall showed significant statistical correlation with A. cajennense infestations on horses. Labruna (2000) has shown under field conditions of Pirassununga that A. cajennense larvae that hatched in early and late February 2000 (from engorged females dropped from the host in pasture during late December 1999 and early January 2000), did not show host seeking activity until late April 2000, when daylight length had dropped to less than 12 h and mean soil temperatures had dropped to below 23 ◦ C. During all larval activity periods (until the following August when all unfed larvae had died), daylight length and mean soil temperature were always below 12 h and 23 ◦ C, respectively. Before late April, daylight length was longer than 12 h and mean soil temperature was always higher than 23 ◦ C. During this time unfed larvae were restricted to the soil base under the vegetation where they remained without host seeking behavior. This free-living stage behavior lasted for 9–11 weeks, as these larvae had hatched in February 2000. Larvae climbed to the top of vegetation (host seeking behavior) in late April, a time that coincided with temperature decreases, end of the rainy season, and shorter daylight periods. These results suggested that the 1-year generation pattern of A. cajennense in the southeastern region of Brazil is controlled primarily by a behavioral diapause of unfed larvae in the pastures during summer months. In fact, it has been reported that tick species with one generation per year tend to have a summer period of dormancy or diapause of unfed ticks (Belozerov, 1982). There was always a greater number of A. cajennense adult males than females on the horses during the 24-month period of observation. This same pattern was also observed by Oliveira (1998) on horses at Pedro Leopoldo, state of Minas Gerais. Laboratory studies have shown that both A. cajennense populations (Pirassununga and Pedro Leopoldo) were sexually comprised in the adult stage by 60–64% females and 30–36% males (Freitas et al., 1999; Pinter et al., 2002). As both the present study and that of Oliveira (1998) used similar methods to quantify the A. cajennense adult stage every 14 days on horses, the most reasonable explanation for the contradiction between sex ratio for field and laboratory collected data is that males may have a much longer parasitic period on horses that resulted in
74
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
the same individual male counted on more than one observation. In fact, Pinter et al. (2002) have shown that A. cajennense males can survive and maintain sexual activity for more than 80 days on laboratory hosts. However, A. cajennense females complete their feeding period within 14 days, when males were present on the host. Therefore, even though the tick adult population was constituted by more females than males, tick assessment on horses every 14 days would rarely count the same individual female, but an individual male could be counted several times, resulting in a greater proportion of males over females on the horse under field conditions. The prolonged parasitic period of A. cajennense males on the host could have important implications making the host more attractive for host seeking conspecific females (Pinter et al., 2002). In the beginning of the adult season (spring), the proportion of males over females on the horses was about twice as high as the proportion at the second half of the season (summer). A similar pattern was observed in the study of Oliveira (1998), in Pedro Leopoldo, and it is possibly related to aggregation-attachment pheromones secreted by males. Rechav et al. (1997) showed evidence that attached A. cajennense males secreted pheromones that significantly increased the number of unfed conspecific male and female ticks that attached to the host. Under field conditions, although a larger proportion of A. cajennense nymphs are molting to female ticks (Pinter et al., 2002), a greater proportion of unfed male ticks might seek for a host earlier than females, improving host seeking behavior of unfed females. In both the present study and that of Oliveira (1998), the horses may be more attractive to male than female ticks in the beginning of the season. As greater amounts of aggregation-attachment pheromones were progressively produced by fed males, female host seeking behavior increased, decreasing the proportion of males over females on the horses during the second half of the season. Although the horses were placed on pasture during September 1997, we only found A. nitens females >4 mm during late November 1997. Cattle removed from the pasture 4 weeks before starting the study were not infested by A. nitens. The horses were infested by A. nitens when they were introduced in the pasture. Therefore, it is probable that the A. nitens ticks found on horses during November 1997 with a peak in December 1997 were the first generation of ticks that originated from engorged females dropped from the horses right after they were introduced to the pasture in September 1997. After this first peak of infestation, there were other consecutive distinct peaks until October–November 1998, which would suggest different generations of A. nitens. In the second year of the study (from November 1998 onwards), mean infestation values were maintained below 100 ticks per horse until the end of the study. This change in pattern was probably the result of removing half of the horses from the pasture at the beginning of the second year (October 1998). The lower stocking rate contributed to diminished chances of unfed A. nitens larvae to encounter a host, and provided larger amounts of forage available for the remaining five horses in the pasture, which improved their nutritional status. In contrast, the decrease of horse density did not diminish the abundance of A. cajennense on the horses in the second year. However, A. cajennense is a three-host tick that completes only one generation a year and its ecology cannot be compared to that of A. nitens, which is a one-host tick that completes more than one generation per year. The highest peaks of A. nitens in the first year were significantly associated with greater decreases of horse PCV values. During the second year, when infestations by A. nitens
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
75
were always low, the mean PCV values were always above the minimum normal reference value (24%) reported for horses (Ferreira-Neto, 1978). Decreasing PCV values on cattle have also been associated with higher infestations by B. microplus, another one-host tick (O’kelly and Seifert, 1970; Kasai et al., 2000). There was no compensatory decrease in PCV values associated with infestation by A. cajennense, even when there were higher infestations by adult ticks on the horses during the summer months. In the present study, the ears and the perineum, tail and groin areas were equally infested with A. nitens. Falce (1986) and Borges et al. (2000) reported that the ear was the predilection site of infestations with A. nitens, since approximately 60% of all ticks were counted in the area of the ear. In the study of Borges et al. (2000), the mane was the second most infested body area (21% of the ticks), however, the mane accounted for only 15% of the ticks counted in the present study. These differences are possibly a result of different predilection sites between tick populations. They might also be affected by horse hair coat conditions as we have observed that horses with longer mane hair tended to have higher tick burdens on this area (data not shown). In the first 3 months of the present study, we found a limited number of B. microplus ticks on the horses. After this period, this tick was not found for the remainder of the study. This reflects that horses are secondary hosts for B. microplus. As long as the primary host (cattle) of B. microplus were in the pasture, infestation by B. microplus would be continually found on other host species (secondary hosts). When cattle were removed from the pasture, the B. microplus population was not able to survive with only horses as hosts. This statement is supported by the fact that infestations by B. microplus on several host species other than cattle have been reported from areas where cattle have been grazing (Moreno, 1984; Bittencourt et al., 1990; Ito et al., 1998; Heuchert et al., 1999; Labruna et al., 2001). The present study quantified the parasitic stages of A. cajennense on horses over two tick generations. During peaks of infestations by immature ticks, we counted more than 10,000 engorged larvae or more than 3000 engorged nymphs on some individual horses (data not shown). Obviously, the total of larvae and nymphs attached to horses could have been much higher, as our counting method considered only full engorged immature ticks attached to horses. During peaks of infestation by adult ticks, more than 1400 ticks were counted on some individual horses. Certainly, these high tick burdens are indicative of horses as primary hosts for all parasitic stages of A. cajennense. Other primary hosts for A. cajennense, such as tapirs Tapirus terrestris and capybaras Hydrochaeris hydrochaeris (Aragão, 1936; Labruna et al., 2001) were never observed on the pastures during the study. Further studies should evaluate biological, immunological and biochemical aspects of the tick–horse interaction in Latin America. Although A. cajennense and A. nitens are native to the neotropical area (Camicas et al., 1998), horses (Equus caballus) were recently introduced through European colonization after the 15th century (Cartelle, 1988).
Acknowledgements We thank J.A. Metzner, A. Santa-Roza and J.R. Devitto for their technical support. This work was supported by FAPESP (Process No. 95/6938-0).
76
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
References Aragão, H., 1936. Ixodidas brasileiros e de alguns paizes limitrophes. Mem. Inst. Oswaldo Cruz 31, 759–843. Archer, R.K., 1977. Technical methods. In: Archer, R.K., Jeffcott, L.B., Lehmann, H. (Eds.), Comparative Clinical Haematology. Blackwell Scientific Publications, Oxford, pp. 537–610. Belozerov, V.N., 1982. Diapause and biological rhythms in ticks. In: Obenchain, F.D., Galun, R. (Eds.), Physiology of Ticks. Pergamon Press, Oxford, 509 pp. Bittencourt, A.J., Fonseca, A.H., Faccini, J.L.H., Bueno, B.F., 1990. Comportamento do Boophilus microplus (Canestrini, 1887) (Acari) em infestações artificiais e naturais em diferentes hospedeiros. Arq. Univ. Rur. Rio de Janeiro 13, 173–182. Borges, L.M.F., Oliveira, P.R., Ribeiro, M.F.B., 2000. Seasonal dynamics of Anocentor nitens on horses in Brazil. Vet. Parasitol. 89, 165–171. Camicas, J.L., Hervy, J.P., Adam, F., Morel, P.C., 1998. Les tiques du monde (Acarida, Ixodida): Nomenclature, stades décrits, hôtes, répartition. Orstom, Paris, 233 pp. Cartelle, C., 1988. Cavalo: Um brasileiro antigo. Ciência Hoje 8, 27–33. Clifford, C.M., Anastos, G., 1960. The use of chaetotaxy in the identification of larval ticks (Acarina: Ixodidae). J. Parasitol. 46, 567–578. Dias, E., Martins, A.V., 1939. Spotted fever in Brazil. Am. J. Trop. Med. 19, 103–108. Falce, H.C., 1986. Infestações múltiplas por ixod´ıdeos (Acari: Ixodidae) em bovinos e equ´ıdeos no primeiro planalto do Estado do Paraná. Rev. Set. Cienc. Agra. 8, 11–13. Ferreira-Neto, J.M., 1978. Patologia Cl´ınica Veterinária. Rabelo e Brasil, Belo Horizonte, 293 pp. Freitas, C.M.V., Leite, R.C., Oliveira, P.R., Bicalho, K.A., Lisboa, C.M., 1999. Comparação entre alguns parâmetros biológicos de ninfas ingurgitadas e adultos não alimentados de Amblyomma cajennense (Fabricius, 1787) (Acari: Ixodidae). In 10th Seminário Brasileiro de Parasitologia Veterinária. Salvador, Brazil, October 24–28, 1999, p. 83. Heuchert, C.M.S., Giulli, V., Athaide, D.F., Bose, R., Friedhoff, K.T., 1999. Seroepidemiologic studies on Babesia equi and Babesia caballi infections in Brazil. Vet. Parasitol. 85, 1–11. Ito, F.H., Vasconcellos, S.A., Bernardi, F., Nascimento, A.A., Labruna, M.B., Arantes, I.G., 1998. Evidência sorológica de brucelose e leptospirose e parasitismo por Ixod´ıdeos em animais silvestres do Pantanal Sul-Mato-Grossense. Ars Veterinária 14, 302–310. Kasai, N., Labruna, M.B., Pires, A.V., Louvandini, H., Abdalla, A.L., 2000. Dinâmica populacional de Boophilus microplus (Canestrini, 1887) em bovinos leiteiros mantidos em manejo de pastejo rotacionado de capim elefante. Arq. Bras. Med. Vet. Zoot. 52, 453–458. Labruna, M.B., 2000. Aspectos da biologia e epidemiologia dos carrapatos de eqüinos no Estado de São Paulo. Tese de Doutorado em Epidemiologia Experimental e Aplicada a Zoonoses. Universidade de São Paulo, São Paulo, 76 pp. Labruna, M.B., Kerber, C.E., Faccini, J.L.H., De Waal, D.T., Gennari, S.M., 2001. Risk factors to tick infestations and their occurrence on horses in the state of São Paulo, Brazil. Vet. Parasitol. 97, 1–14. Linthicum, K.J., Logan, T.M., Bailey, C.L., Gordon, S.W., Peters, C.J., Monath, T.P., Osorio, J., Francy, D.B., McLean, R.G., Leduc, J.W., Graham, R.R., Jahrling, P.B., Moulton, J.R., Dohm, D.J., 1991. Venezuelan equine Encephalomyelitis virus infection and transmission by the tick Amblyomma cajennense (Arachnida: Ixodidae). J. Med. Entomol. 28, 405–409. Moreno, E.C., 1984. Incidência de ixod´ıdeos em bovinos de leite e prevalência em animais domésticos da Região Metalúrgica de Minas Gerais. Dissertação de mestrado em Medicina Veterinária. Universidade Federal de Minas Gerais, Belo Horizonte, 105 pp. O’kelly, J.C., Seifert, G.W., 1970. The effects of tick Boophilus microplus infestations on the blood composition of Shorthorn X Hereford cattle on high and low planes of nutrition. Aust. J. Biol. Sci. 23, 681–690. Oliveira, P.R., 1998. Amblyomma cajennense (Fabricius, 1787) (Acari: Ixodidae): Avaliação de técnicas para o estudo de dinâmica populacional e biotecnologia. Tese de Doutorado em Medicina Veterinária. Universidade Federal de Minas Gerais, Belo Horizonte, 97 pp. Oliveira, P.R., Borges, L.M.F., Lopes, C.M.L., Leite, R.C., 2000. Population dynamics of the free living stages of Amblyomma cajennense (Fabricius, 1787) (Acari: Ixodidae) on pastures of Pedro Leopoldo, Minas Gerais state, Brazil. Vet. Parasitol. 92, 295–301.
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
77
Pinter, A.S., Labruna, M.B., Faccini, J.L.H., 2002. Feeding period of males and females of Amblyomma cajennense (Acari: Ixodidae) under laboratory conditions, with special reference to sex ratio. Vet. Parasitol., 105, 79–88. Rechav, Y., Goldberg, M., Fielden, L.J., 1997. Evidence for attachment pheromones in the Cayenne tick (Acari: Ixodidae). J. Med. Entomol. 34, 234–237. Roby, T.O., Anthony, D.W., 1963. Transmission of equine piroplasmosis by Dermacentor nitens Neumann. JAVMA 142, 768–769. Serra-Freire, N.M.S., 1982. Epidemiologia de Amblyomma cajennense: ocorrência estacional e comportamento dos estádios não parasitários em pastagens do Estado do Rio de Janeiro. Arq. Univ. Fed. Rur. Rio de Janeiro 5, 187–193. Souza, A.P., 1990. Variação populacional dos principais ixod´ıdeos parasitas de bovinos e eqüinos em diferentes condições de manejo, nos munic´ıpios de Paracambi e Itagua´ı no estado do Rio de Janeiro. Tese de Doutorado em Parasitologia Animal. Universidade Federal Rural do Rio de Janeiro, Seropédica, 81 pp. SPSS for Windows, 1999. SPSS Base 9.0.1. Users Guide. SPSS Inc., Chicago, 740 pp.
Seasonal dynamics of ticks (Acari: Ixodidae) on horses in the state of São Paulo, Brazil Marcelo B. Labruna a,∗ , Nobuko Kasai a , Fernando Ferreira a , João L.H. Faccini b , Solange M. Gennari a a
Faculdade de Medicina Veterinária e Zootecnia, Departamento de Medicina Veterinária Preventiva e Saúde Animal, Universidade de São Paulo, Av. Prof. Orlando Marques de Paiva, 87 Cidade Universitária, São Paulo, SP 05508-000, Brazil b Departamento de Parasitologia Animal, Instituto de Veterinária, Universidade Federal Rural do Rio de Janeiro, Km 7, BR 465, Seropédica, RJ 23890-000, Brazil Received 27 July 2001; received in revised form 20 November 2001; accepted 26 November 2001
Abstract Natural tick infestations were assessed every 14 days on horses over a 2-year period. Amblyomma cajennense adult ticks were counted individually, without detachment from the horses. Larvae and nymphs of A. cajennense were collected using a rubber scraper that scratched engorged immature ticks from the host. Adult females of Anocentor nitens larger than 4 mm length were counted on the horses. Blood samples were also obtained from the horses every 14 days and macroclimatic data were obtained for the study period. Infestations of A. cajennense demonstrated distinct peaks of activity for each of the three parasitic stages over each 12-month period, showing a 1-year generation pattern. Larvae predominated from April to July and nymphs from June to October. Adults predominated from October to March with a greater number of adult males than females. Although other studies on seasonal dynamics in the states of Rio de Janeiro and Minas Gerais were performed with the free-living stages of A. cajennense on pastures, the present study in the state of São Paulo, performed with the parasitic stages of A. cajennense on horses, showed similar results to those observed in other states. Infestations by A. nitens demonstrated distinct peaks of activity of adult females (>4 mm), suggesting different tick generations during the year. Infestation with A. nitens was much higher in the first year than the second year which may have been related to horse nutritional status and stocking rate. Although several climatic variables showed statistical significant correlation (r) with tick counts, the determination coefficients (R2 ) were always lower than 0.40, suggesting that any single significant variable (i.e. mean temperature) would not explain the tick distribution pattern over the year. The highest peaks of A. nitens females (>4 mm) were significantly associated with decrease in horse packed cell volumes (R 2 = 0.603). The ears and
∗ Corresponding author. Tel.: +55-11-3818-7703; fax: +55-11-3818-7028. E-mail address: [email protected] (M.B. Labruna).
0304-4017/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 4 0 1 7 ( 0 1 ) 0 0 6 4 9 - 5
66
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
the perineum, tail and groin region accounted for around 70% of all A. nitens females counted on the horses. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Anocentor nitens; Amblyomma cajennense; Seasonal dynamics; Horse; Brazil
1. Introduction Several studies have reported that three tick species infest horses in Brazil: Anocentor nitens, Amblyomma cajennense and less commonly, Boophilus microplus (Moreno, 1984; Falce, 1986; Heuchert et al., 1999; Labruna et al., 2001). Horses act as primary hosts for A. nitens and A. cajennense, as the establishment of populations of both species in areas inhabited by horses is not dependent on the presence of other host species (Labruna et al., 2001). Conversely, horses are secondary hosts for B. microplus, as the presence of this tick is dependent on the presence of its primary host cattle, grazing with horses in the same area (Moreno, 1984; Heuchert et al., 1999; Labruna et al., 2001). A. nitens is a one-host tick and was shown to be the most common tick species on horses in the state of São Paulo (Labruna et al., 2001). It attacks the host, particularly the region of the ear, nasal diverticulum, mane and perineum (Borges et al., 2000). Infestations can cause severe lesions in the ears that predispose the host to bacterial infections and screw-worm infestations. A. nitens is the natural vector of Babesia caballi in Latin America (Roby and Anthony, 1963). A. cajennense, a three-host tick, is also implicated as a serious ectoparasite of horses and is the principal tick infesting humans in central and southern Brazil (Aragão, 1936; Labruna et al., 2001). A. cajennense is the main vector of Rickettsia rickettsii, the agent of human Brazilian spotted fever (Dias and Martins, 1939). It was also proven to be a competent vector of the Venezuelan equine encephalomyelitis virus under laboratory conditions (Linthicum et al., 1991). In a 1-year period of field sampling of unfed larvae, nymphs and adults of A. cajennense in horse pastures in the state of Rio de Janeiro (Serra-Freire, 1982), it was suggested that this tick completed one generation a year. This proposition was later confirmed in the studies of Souza (1990) in the state of Rio de Janeiro and Oliveira et al. (2000) in the state of Minas Gerais. Studies regarding the seasonal dynamics of A. nitens are limited. Souza (1990) and Borges et al. (2000) reported 3–4 peaks of host seeking larvae on the pasture or adult females on the horses during the year, throughout a 2-year study in the states of Rio de Janeiro and Minas Gerais, respectively. They suggested that A. nitens completed 3–4 generations per year in the studied areas. This paper reports the seasonal dynamics of tick infestations on horses over a 2-year period in the state of São Paulo. 2. Materials and methods 2.1. Study site and horses This study was conducted at the experimental ranch of the Veterinary Faculty of the University of São Paulo, located at Pirassununga County at 631 m of altitude (21◦ 59 24 S,
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
67
47◦ 25 12 W), state of São Paulo, Brazil. On September 5th, 1997, 10 crossbred horses, 3–7 years of age (mean: 4.5 years old), were kept in a 4 ha pasture that was naturally infested with A. cajennense and B. microplus, based on previous examination of cattle that occupied the pasture and were removed 4 weeks before the introduction of the study horses. The horses came from a commercial farm located at Pirassununga and were naturally infested with A. nitens at the time they were placed on the pasture. Pasture consisted of forage grasses (Paspalum notatum, Brachiaria decumbens) and also several bushes and shrub species (Acacia bonariensis, Ageratum conyzoides, Senna occidentalis and Sida spp). 2.2. Tick counts Evaluation of natural tick infestations on the horses started on October 15th, 1997 and ended on September 29th, 1999. A. nitens tick counts were continued for two additional months, until November 24th, 1999. Horses were never treated with acaricides during the experiment. On October 14th, 1998, five horses were endangered by high tick burdens and low packed cell volume (PCV) levels. These five horses were removed from the pasture. Therefore, the pasture was managed with 2.5 horses/ha in the first year and 1.25 horse/ha in the second year. During the experiment tick burdens on the horses were evaluated every 14 days, always on Wednesday mornings, totaling 52 tick count observations during the study for A. cajennense and 56 tick count observations for A. nitens. Counts of A. cajennense were performed according to the methodology described and tested by Oliveira (1998) for the three parasitic stages of A. cajennense on horses. Ticks were evaluated on the entire left side of the horse body and the total number of ticks was doubled to estimate the total number of ticks on each animal. Adult males and females of A. cajennense were counted individually, without detaching the ticks. Larvae and nymphs of A. cajennense were checked using a rubber scraper which scratched engorged immature ticks from the host body. This procedure removed only totally engorged immature ticks (Oliveira, 1998). Prior to counting the number of engorged immature ticks, they were maintained in the laboratory at 27 ◦ C for 7 days to allow engorged larvae of A. nitens to molt to nymphs. During the first 8 months of the experiment, all engorged larvae that died before molting were identified to generic level (Clifford and Anastos, 1960). Thus, larvae identified in the genus Amblyomma were considered to be A. cajennense, whereas larvae identified in the genus Dermacentor or Anocentor were considered to be A. nitens. A total of 1248 dead larvae were examined during the 8 months that this procedure was conducted. As only eight (0.6%) larvae belonged to the genus Anocentor, this procedure was discontinued and all engorged dead larvae were counted as A. cajennense. Although the differentiation of engorged nymphs of both species was easily performed by macroscopic visualization, the 7 day-period that the ticks were maintained at 27 ◦ C was also sufficient for engorged nymphs of A. nitens to molt to adults. Evaluation of A. nitens on the horses, was by the method described by Borges et al. (2000). Female ticks larger than 4 mm length were counted on the right side of each horse. For this purpose, the entire right side of the horse was divided into five areas which were designated as: (1) ear; (2) mane; (3) face and neck; (4) perineum, tail and groin; (5) thorax, ventrum and legs. Ticks present in the nasal diverticulum were not counted due to the difficulty of
68
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
assessing this anatomic area. The number of ticks in each of the five areas was doubled and summed to estimate the total number of ticks on each animal. Blood samples were obtained through the jugular vein into individual EDTA-coated glass tubes every 14 days for determination of the PCV values by the microhematocrit method (Archer, 1977). During the study period all horses were orally drenched with Oxibendazol (Equitac, Pfizer) every 2 or 3 months for control of helminth parasitism. 2.3. Environmental data Macroclimatic data for the study site were obtained from the meteorological station of the experimental farm, situated approximately 150 m from the pasture where the horses were maintained during the experiment. Daily measurements of maximum, minimum and median temperatures, total rainfall and median and minimum relative humidities (RHs) were obtained for the entire study period. Data on temperature and RH are presented as the mean of the previous 14 days to each tick count and data on rainfall as the sum of the previous 14 days for each tick count. Daylight length (photoperiod) for the study site over the study period was obtained from a Geo Position System electronic device (Carmin GPS 12 XL, Olathe, KS, USA). 2.4. Data analysis Scatter plots between the mean tick counts and climatic variables were performed to verify if there were any linear associations between the variables. Pearson correlation coefficients (r) were calculated for each of the climatic, PCVs and mean tick count values for the entire study period. All analyses were performed using SPSS for Windows (1999).
3. Results Counts of A. cajennense on horses demonstrated distinct peaks of activity for each of the three parasitic stages over each 12-month period (Fig. 1). Larval peaks were observed from April to July in both years, when 250,850 engorged larvae (98.2% of the total larvae) were counted. The highest larval peaks occurred in May when mean tick counts were 5645±2151 larvae per horse in the first year and 5798 ± 3802 per horse in the second year (Fig. 1). From October to February of both years, no A. cajennense larvae were found on horses. Nymphal peaks occurred from June to October in both years when 34,794 engorged nymphs (99.5% of the total nymphs) were counted. The highest nymphal peaks occurred in August and September when mean tick counts were 406.4 ± 648 nymphs per host in the first year and 850.4 ± 1144 in the second year. No engorged nymph was found on horses from January to April in the first year and only 12 engorged nymphs were counted during this same period in the second year. Peaks of A. cajennense adult ticks occurred from October to March when 54,556 ticks (83.7% of the total adults) were counted and the mean tick counts were always higher than 100 adult ticks per horse in the first year and higher than 200 in the second year. The highest peak of A. cajennense adult ticks on horses occurred in February 1999 when a mean of 1053 ± 413.5 adult ticks per horse was found. Despite showing distinct peaks from
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
69
Fig. 1. Seasonal dynamics of A. cajennense on horses pastured on a ranch in Pirassununga, São Paulo. Infestations by adults (A), larvae (B) and nymphs (C) were evaluated on the horses every 14 days from October 1997 to September 1999. In graph A, the plotted line represents the mean number of adults on the horses and the bars represent the mean ratio of the number of adult males over females on the horses.
70
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
Fig. 2. Climatic data at Pirassununga County from September 1997 to November 1999.
September to March in both years, the A. cajennense adult stage was found on horses at all count dates of the study. However, adult tick numbers were very low and at many times near zero in counts performed at times other than September–March (Fig. 1). Comparisons between climatic data obtained for the study period (Fig. 2) and tick count data presented in Fig. 1 showed that the periods of highest activities of A. cajennense larvae and nymphs were the autumn and winter months, when there were shorter daylight periods and lower mean temperatures, RH and rainfall. On the other hand, the periods of highest activity of the A. cajennense adult stage occurred during the spring and summer months when there were longer daylight periods and higher mean temperatures, RH and rainfall. The correlation coefficients between the mean tick counts and the climatic data for the study period are shown in Table 1. Engorged larvae counts showed significant negative correlation (P < 0.05) with the mean values of maximum, minimum and mean temperatures, and daylight hours. Similarly, engorged nymph counts showed significant
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
71
Table 1 Results of statistical correlation analysis between the climatic variables and tick count values on horses over a 24-month period in Pirassununga, São Paulo (data are presented as the Pearson correlation coefficient (r) followed by the P value in parentheses) Climatic variables
Maximum temperature Minimum temperature Mean temperature Mean RH Minimum RH Total rainfall Photoperiod (light hours)
A. cajennense Larva
Nymph
Adult
−0.52 (<0.001) −0.55 (<0.001) −0.50 (<0.001) 0.04 (0.741) −0.16 (0.253) −0.25 (0.070) −0.56 (<0.001)
−0.22 (0.113) −0.38 (0.005) −0.51 (<0.001) −0.60 (<0.001) −0.62 (<0.001) −0.35 (0.100) −0.39 (0.004)
0.46 (0.001) 0.51 (<0.001) 0.56 (<0.001) 0.26 (0.054) 0.31 (0.022) 0.47 (<0.001) 0.55 (<0.001)
A. nitens (females >4 mm) −0.44 (0.001) −0.39 (0.003) −0.33 (0.011) 0.14 (0.290) 0.12 (0.928) −0.12 (0.456) −0.37 (0.004)
negative correlation (P < 0.05) with the mean values of minimum and mean temperatures, minimum and mean RH, and daylight hours. On the other hand, the adult counts showed significant positive correlation (P < 0.05) with the mean values of maximum, minimum and mean temperatures, minimum RH, total rainfall and daylight hours. The scatter plots indicated that none of the climatic variables presented a linear relationship with the A. cajennense counts, thus making unnecessary the application of the linear multivariable model. Although there were statistically significant correlations between the climatic data and the mean values of A. cajennense larvae, nymphs and adults on horses, the determination coefficients (R2 ) were always low (<0.40). In this case, R2 indicates the proportion of the tick count variance that is attributable to variation in the fitted mean of tick count values, as a function of the climatic data. The results suggested that any single significant variable alone (i.e. mean temperature) would not explain the tick distribution pattern over the year. Infestations by adult A. cajennense on the horses were characterized by a higher number of male than female ticks during all of the study period (Fig. 1A). During the period of highest adult activity (October–March) the proportion of males over females was higher in the first half of this period (October–December) than the second half (January–March). In the first year, this proportion was approximately four males for each female [4:1 (M:F)] from October to December 1997, dropping to about 2:1 (M:F) in the period from January to March 1998. Similarly, in the second year this proportion was about 2.5:1 (M:F) from October to December 1998 dropping to about 1.4:1 (M:F) during January–March 1999. The proportions of males to females during the rest of the year did not show a definitive pattern. These periods were characterized by the highest activity of immature ticks and few adult ticks were present on horses, some times only male ticks were observed (data not shown). The parasitism by A. nitens engorged females >4 mm was first recorded on the horses by the end of November 1997, 2 months after tick counts began (Fig. 3). For this reason, the A. nitens counts were continued until November 1999. The infestation pattern by A. nitens was characterized by different peaks over the year. The first peak of infestation occurred during December 1997, and after that, four other peaks through November 1998. In the first year, the highest peaks of A. nitens infestation occurred in April 1998 (mean: 273.3 ± 179.2 ticks per horse) and July 1998 (291.4 ± 196.4). A third peak of infestation occurred in October 1998 (219.0 ± 178.4). From November 1998 to November 1999, when the stocking rate
72
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
Fig. 3. Mean number of A. nitens females >4 mm in length (•) and mean PCV (䊏) evaluated on horses every 14 days from October 1997 to November 1999 at Pirassununga, São Paulo.
of the pasture had dropped to half of the first year, A. nitens counts were much lower as all counts showed mean values lower than 100 ticks per horse. Therefore, similar to the first year, the highest peaks of infestation in the second year occurred during the first semester followed by a drastic reduction of infestation in August 1999 and new higher peaks of infestation during September–October 1999. Although there were statistically significant negative correlations between the mean, maximum and minimum temperatures and photoperiod data with mean values of A. nitens on horses, the determination coefficients (R2 ) were also always low (<0.40). As the scatter plots did not indicate any linear relationship between the climatic variables presented in Fig. 2 and A. nitens counts, the application of a linear multivariable model was again unnecessary. The mean values of PCV over the study showed significant negative statistical correlation with A. nitens count values (P < 0.001), showing a R 2 = −0.603. This coefficient indicated that the A. nitens infestation explained 60.3% of the PCV variation observed on the horses. In fact, it can be seen in Fig. 3 that the higher A. nitens peaks coincided in time with more pronounced decreases of PCV mean values. The lowest mean PCV value (23.4±5.0%) was observed on July 1998 when the highest mean A. nitens count (291.4 ± 196.4) was also observed. The predilection sites for A. nitens infestation on the horse were the ears and the perineum, tail and groin region. From a total of 13,730 adult females >4 mm counted on the right side of the horses, 5086 (37.0%) were in the internal face of the right ear and 4414 (32.2%) in the perineum, tail and groin area. Mane and head accounted for 14.6 and 12.0% of the total ticks, respectively. Only 4.2% of the ticks were counted on thorax, ventrum and legs. During the first six tick counts of the study, from October to December 1997, an infestation of at least one horse with B. microplus was noted at each tick count. B. microplus infestation
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
73
was low with no more than six B. microplus adult ticks on each horse. After December 1997, no B. microplus ticks were observed on the horses for the remainder of the study.
4. Discussion Oliveira (1998) determined that infestations by the three parasitic stages of A. cajennense on horses showed a marked symmetry, making it possible to obtain an adequate estimate of population by counting only one side of the host. In addition, Oliveira (1998) determined that there was no significant difference between counting ticks for every 7 or 14 days. The A. cajennense population of São Paulo state demonstrated a 1-year generation pattern. Although other studies on seasonal dynamics in the states of Rio de Janeiro (Serra-Freire, 1982; Souza, 1990) and Minas Gerais (Oliveira et al., 2000) were performed with the free-living stages of A. cajennense on the pastures, the present study was concentrated on the parasitic stages of A. cajennense on horses and provided similar results to those observed in other states. In both studies, immature ticks predominated in the autumn/winter months and the adult stage predominated in the spring/summer months. Photoperiod and climatic variables such as temperature, RH and rainfall showed significant statistical correlation with A. cajennense infestations on horses. Labruna (2000) has shown under field conditions of Pirassununga that A. cajennense larvae that hatched in early and late February 2000 (from engorged females dropped from the host in pasture during late December 1999 and early January 2000), did not show host seeking activity until late April 2000, when daylight length had dropped to less than 12 h and mean soil temperatures had dropped to below 23 ◦ C. During all larval activity periods (until the following August when all unfed larvae had died), daylight length and mean soil temperature were always below 12 h and 23 ◦ C, respectively. Before late April, daylight length was longer than 12 h and mean soil temperature was always higher than 23 ◦ C. During this time unfed larvae were restricted to the soil base under the vegetation where they remained without host seeking behavior. This free-living stage behavior lasted for 9–11 weeks, as these larvae had hatched in February 2000. Larvae climbed to the top of vegetation (host seeking behavior) in late April, a time that coincided with temperature decreases, end of the rainy season, and shorter daylight periods. These results suggested that the 1-year generation pattern of A. cajennense in the southeastern region of Brazil is controlled primarily by a behavioral diapause of unfed larvae in the pastures during summer months. In fact, it has been reported that tick species with one generation per year tend to have a summer period of dormancy or diapause of unfed ticks (Belozerov, 1982). There was always a greater number of A. cajennense adult males than females on the horses during the 24-month period of observation. This same pattern was also observed by Oliveira (1998) on horses at Pedro Leopoldo, state of Minas Gerais. Laboratory studies have shown that both A. cajennense populations (Pirassununga and Pedro Leopoldo) were sexually comprised in the adult stage by 60–64% females and 30–36% males (Freitas et al., 1999; Pinter et al., 2002). As both the present study and that of Oliveira (1998) used similar methods to quantify the A. cajennense adult stage every 14 days on horses, the most reasonable explanation for the contradiction between sex ratio for field and laboratory collected data is that males may have a much longer parasitic period on horses that resulted in
74
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
the same individual male counted on more than one observation. In fact, Pinter et al. (2002) have shown that A. cajennense males can survive and maintain sexual activity for more than 80 days on laboratory hosts. However, A. cajennense females complete their feeding period within 14 days, when males were present on the host. Therefore, even though the tick adult population was constituted by more females than males, tick assessment on horses every 14 days would rarely count the same individual female, but an individual male could be counted several times, resulting in a greater proportion of males over females on the horse under field conditions. The prolonged parasitic period of A. cajennense males on the host could have important implications making the host more attractive for host seeking conspecific females (Pinter et al., 2002). In the beginning of the adult season (spring), the proportion of males over females on the horses was about twice as high as the proportion at the second half of the season (summer). A similar pattern was observed in the study of Oliveira (1998), in Pedro Leopoldo, and it is possibly related to aggregation-attachment pheromones secreted by males. Rechav et al. (1997) showed evidence that attached A. cajennense males secreted pheromones that significantly increased the number of unfed conspecific male and female ticks that attached to the host. Under field conditions, although a larger proportion of A. cajennense nymphs are molting to female ticks (Pinter et al., 2002), a greater proportion of unfed male ticks might seek for a host earlier than females, improving host seeking behavior of unfed females. In both the present study and that of Oliveira (1998), the horses may be more attractive to male than female ticks in the beginning of the season. As greater amounts of aggregation-attachment pheromones were progressively produced by fed males, female host seeking behavior increased, decreasing the proportion of males over females on the horses during the second half of the season. Although the horses were placed on pasture during September 1997, we only found A. nitens females >4 mm during late November 1997. Cattle removed from the pasture 4 weeks before starting the study were not infested by A. nitens. The horses were infested by A. nitens when they were introduced in the pasture. Therefore, it is probable that the A. nitens ticks found on horses during November 1997 with a peak in December 1997 were the first generation of ticks that originated from engorged females dropped from the horses right after they were introduced to the pasture in September 1997. After this first peak of infestation, there were other consecutive distinct peaks until October–November 1998, which would suggest different generations of A. nitens. In the second year of the study (from November 1998 onwards), mean infestation values were maintained below 100 ticks per horse until the end of the study. This change in pattern was probably the result of removing half of the horses from the pasture at the beginning of the second year (October 1998). The lower stocking rate contributed to diminished chances of unfed A. nitens larvae to encounter a host, and provided larger amounts of forage available for the remaining five horses in the pasture, which improved their nutritional status. In contrast, the decrease of horse density did not diminish the abundance of A. cajennense on the horses in the second year. However, A. cajennense is a three-host tick that completes only one generation a year and its ecology cannot be compared to that of A. nitens, which is a one-host tick that completes more than one generation per year. The highest peaks of A. nitens in the first year were significantly associated with greater decreases of horse PCV values. During the second year, when infestations by A. nitens
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
75
were always low, the mean PCV values were always above the minimum normal reference value (24%) reported for horses (Ferreira-Neto, 1978). Decreasing PCV values on cattle have also been associated with higher infestations by B. microplus, another one-host tick (O’kelly and Seifert, 1970; Kasai et al., 2000). There was no compensatory decrease in PCV values associated with infestation by A. cajennense, even when there were higher infestations by adult ticks on the horses during the summer months. In the present study, the ears and the perineum, tail and groin areas were equally infested with A. nitens. Falce (1986) and Borges et al. (2000) reported that the ear was the predilection site of infestations with A. nitens, since approximately 60% of all ticks were counted in the area of the ear. In the study of Borges et al. (2000), the mane was the second most infested body area (21% of the ticks), however, the mane accounted for only 15% of the ticks counted in the present study. These differences are possibly a result of different predilection sites between tick populations. They might also be affected by horse hair coat conditions as we have observed that horses with longer mane hair tended to have higher tick burdens on this area (data not shown). In the first 3 months of the present study, we found a limited number of B. microplus ticks on the horses. After this period, this tick was not found for the remainder of the study. This reflects that horses are secondary hosts for B. microplus. As long as the primary host (cattle) of B. microplus were in the pasture, infestation by B. microplus would be continually found on other host species (secondary hosts). When cattle were removed from the pasture, the B. microplus population was not able to survive with only horses as hosts. This statement is supported by the fact that infestations by B. microplus on several host species other than cattle have been reported from areas where cattle have been grazing (Moreno, 1984; Bittencourt et al., 1990; Ito et al., 1998; Heuchert et al., 1999; Labruna et al., 2001). The present study quantified the parasitic stages of A. cajennense on horses over two tick generations. During peaks of infestations by immature ticks, we counted more than 10,000 engorged larvae or more than 3000 engorged nymphs on some individual horses (data not shown). Obviously, the total of larvae and nymphs attached to horses could have been much higher, as our counting method considered only full engorged immature ticks attached to horses. During peaks of infestation by adult ticks, more than 1400 ticks were counted on some individual horses. Certainly, these high tick burdens are indicative of horses as primary hosts for all parasitic stages of A. cajennense. Other primary hosts for A. cajennense, such as tapirs Tapirus terrestris and capybaras Hydrochaeris hydrochaeris (Aragão, 1936; Labruna et al., 2001) were never observed on the pastures during the study. Further studies should evaluate biological, immunological and biochemical aspects of the tick–horse interaction in Latin America. Although A. cajennense and A. nitens are native to the neotropical area (Camicas et al., 1998), horses (Equus caballus) were recently introduced through European colonization after the 15th century (Cartelle, 1988).
Acknowledgements We thank J.A. Metzner, A. Santa-Roza and J.R. Devitto for their technical support. This work was supported by FAPESP (Process No. 95/6938-0).
76
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
References Aragão, H., 1936. Ixodidas brasileiros e de alguns paizes limitrophes. Mem. Inst. Oswaldo Cruz 31, 759–843. Archer, R.K., 1977. Technical methods. In: Archer, R.K., Jeffcott, L.B., Lehmann, H. (Eds.), Comparative Clinical Haematology. Blackwell Scientific Publications, Oxford, pp. 537–610. Belozerov, V.N., 1982. Diapause and biological rhythms in ticks. In: Obenchain, F.D., Galun, R. (Eds.), Physiology of Ticks. Pergamon Press, Oxford, 509 pp. Bittencourt, A.J., Fonseca, A.H., Faccini, J.L.H., Bueno, B.F., 1990. Comportamento do Boophilus microplus (Canestrini, 1887) (Acari) em infestações artificiais e naturais em diferentes hospedeiros. Arq. Univ. Rur. Rio de Janeiro 13, 173–182. Borges, L.M.F., Oliveira, P.R., Ribeiro, M.F.B., 2000. Seasonal dynamics of Anocentor nitens on horses in Brazil. Vet. Parasitol. 89, 165–171. Camicas, J.L., Hervy, J.P., Adam, F., Morel, P.C., 1998. Les tiques du monde (Acarida, Ixodida): Nomenclature, stades décrits, hôtes, répartition. Orstom, Paris, 233 pp. Cartelle, C., 1988. Cavalo: Um brasileiro antigo. Ciência Hoje 8, 27–33. Clifford, C.M., Anastos, G., 1960. The use of chaetotaxy in the identification of larval ticks (Acarina: Ixodidae). J. Parasitol. 46, 567–578. Dias, E., Martins, A.V., 1939. Spotted fever in Brazil. Am. J. Trop. Med. 19, 103–108. Falce, H.C., 1986. Infestações múltiplas por ixod´ıdeos (Acari: Ixodidae) em bovinos e equ´ıdeos no primeiro planalto do Estado do Paraná. Rev. Set. Cienc. Agra. 8, 11–13. Ferreira-Neto, J.M., 1978. Patologia Cl´ınica Veterinária. Rabelo e Brasil, Belo Horizonte, 293 pp. Freitas, C.M.V., Leite, R.C., Oliveira, P.R., Bicalho, K.A., Lisboa, C.M., 1999. Comparação entre alguns parâmetros biológicos de ninfas ingurgitadas e adultos não alimentados de Amblyomma cajennense (Fabricius, 1787) (Acari: Ixodidae). In 10th Seminário Brasileiro de Parasitologia Veterinária. Salvador, Brazil, October 24–28, 1999, p. 83. Heuchert, C.M.S., Giulli, V., Athaide, D.F., Bose, R., Friedhoff, K.T., 1999. Seroepidemiologic studies on Babesia equi and Babesia caballi infections in Brazil. Vet. Parasitol. 85, 1–11. Ito, F.H., Vasconcellos, S.A., Bernardi, F., Nascimento, A.A., Labruna, M.B., Arantes, I.G., 1998. Evidência sorológica de brucelose e leptospirose e parasitismo por Ixod´ıdeos em animais silvestres do Pantanal Sul-Mato-Grossense. Ars Veterinária 14, 302–310. Kasai, N., Labruna, M.B., Pires, A.V., Louvandini, H., Abdalla, A.L., 2000. Dinâmica populacional de Boophilus microplus (Canestrini, 1887) em bovinos leiteiros mantidos em manejo de pastejo rotacionado de capim elefante. Arq. Bras. Med. Vet. Zoot. 52, 453–458. Labruna, M.B., 2000. Aspectos da biologia e epidemiologia dos carrapatos de eqüinos no Estado de São Paulo. Tese de Doutorado em Epidemiologia Experimental e Aplicada a Zoonoses. Universidade de São Paulo, São Paulo, 76 pp. Labruna, M.B., Kerber, C.E., Faccini, J.L.H., De Waal, D.T., Gennari, S.M., 2001. Risk factors to tick infestations and their occurrence on horses in the state of São Paulo, Brazil. Vet. Parasitol. 97, 1–14. Linthicum, K.J., Logan, T.M., Bailey, C.L., Gordon, S.W., Peters, C.J., Monath, T.P., Osorio, J., Francy, D.B., McLean, R.G., Leduc, J.W., Graham, R.R., Jahrling, P.B., Moulton, J.R., Dohm, D.J., 1991. Venezuelan equine Encephalomyelitis virus infection and transmission by the tick Amblyomma cajennense (Arachnida: Ixodidae). J. Med. Entomol. 28, 405–409. Moreno, E.C., 1984. Incidência de ixod´ıdeos em bovinos de leite e prevalência em animais domésticos da Região Metalúrgica de Minas Gerais. Dissertação de mestrado em Medicina Veterinária. Universidade Federal de Minas Gerais, Belo Horizonte, 105 pp. O’kelly, J.C., Seifert, G.W., 1970. The effects of tick Boophilus microplus infestations on the blood composition of Shorthorn X Hereford cattle on high and low planes of nutrition. Aust. J. Biol. Sci. 23, 681–690. Oliveira, P.R., 1998. Amblyomma cajennense (Fabricius, 1787) (Acari: Ixodidae): Avaliação de técnicas para o estudo de dinâmica populacional e biotecnologia. Tese de Doutorado em Medicina Veterinária. Universidade Federal de Minas Gerais, Belo Horizonte, 97 pp. Oliveira, P.R., Borges, L.M.F., Lopes, C.M.L., Leite, R.C., 2000. Population dynamics of the free living stages of Amblyomma cajennense (Fabricius, 1787) (Acari: Ixodidae) on pastures of Pedro Leopoldo, Minas Gerais state, Brazil. Vet. Parasitol. 92, 295–301.
M.B. Labruna et al. / Veterinary Parasitology 105 (2002) 65–77
77
Pinter, A.S., Labruna, M.B., Faccini, J.L.H., 2002. Feeding period of males and females of Amblyomma cajennense (Acari: Ixodidae) under laboratory conditions, with special reference to sex ratio. Vet. Parasitol., 105, 79–88. Rechav, Y., Goldberg, M., Fielden, L.J., 1997. Evidence for attachment pheromones in the Cayenne tick (Acari: Ixodidae). J. Med. Entomol. 34, 234–237. Roby, T.O., Anthony, D.W., 1963. Transmission of equine piroplasmosis by Dermacentor nitens Neumann. JAVMA 142, 768–769. Serra-Freire, N.M.S., 1982. Epidemiologia de Amblyomma cajennense: ocorrência estacional e comportamento dos estádios não parasitários em pastagens do Estado do Rio de Janeiro. Arq. Univ. Fed. Rur. Rio de Janeiro 5, 187–193. Souza, A.P., 1990. Variação populacional dos principais ixod´ıdeos parasitas de bovinos e eqüinos em diferentes condições de manejo, nos munic´ıpios de Paracambi e Itagua´ı no estado do Rio de Janeiro. Tese de Doutorado em Parasitologia Animal. Universidade Federal Rural do Rio de Janeiro, Seropédica, 81 pp. SPSS for Windows, 1999. SPSS Base 9.0.1. Users Guide. SPSS Inc., Chicago, 740 pp.