Sequential Histopathology at the Rhipicephalus ... - Springer Link

Experimental and Applied Acarology 23: 915–928, 1999. © 1999 Kluwer Academic Publishers. Printed in the Netherlands.

Sequential histopathology at the Rhipicephalus sanguineus tick feeding site on dogs and guinea pigs ´ and G.H. BECHARA M.P.J. SZABO* Departamento de Patologia Veterinária, Faculdade de Ciências Agrárias e Veterinárias, Universidade Estadual Paulista, 14.870–000. Jaboticabal, SP, Brasil (Received 19 October 1998; revised 8 April 1999; accepted 21 July 1999)

Abstract. The tick Rhipicephalus sanguineus is a very common parasite of dogs worldwide. Dogs seem unable to acquire resistance against this tick species, whereas guinea pigs demonstrate a very strong resistance following primary infestation. We studied the inflammatory reaction at the R. sanguineus tick feeding site on dogs and guinea pigs during primary and tertiary infestations at different time intervals after attachment. Biopsies were collected after 4, 24, 48 and 96 hours. Changes that were found in all experimental groups included a cone of cement around the mouthparts of the tick, epidermal hyperplasia, edema and inflammatory cell infiltration in the dermis directly underneath the tick attachment site. Dogs reacted to ticks mainly with neutrophils, particularly after repeated exposure. Mast cells and mononuclear leukocytes were also present. Guinea pigs reacted to R. sanguineus mainly with mononuclear cells, eosinophils and basophils. These cells were particularly numerous after repeated exposure to R. sanguineus. Our results suggest that basophils and eosinophils are involved in resistance of guinea pigs to R. sanguineus and that neutrophils in dogs have little effect against this tick species. Key words: histopathology, leukocytes, dog, guinea pig, resistance, tick, Rhipicephalus sanguineus

Introduction Several hosts develop resistance to ticks after repeated infestations as reviewed by Rechav (1992) and Wikel (1996). Usually laboratory animal-tick associations are characterized by a more intense expression of acquired resistance than animal-tick relationships that occur naturally (Ribeiro, 1989). In fact, many studies involving naturally occurring host-tick relationships were unable to detect resistance of hosts to ticks even after repeated infestations (Randolph, 1979; Fielden et al., 1992). Rhipicephalus sanguineus (Latreille, 1806) is probably the most widely distributed tick species in the world (Pegram et al., 1987) and has been linked to tick-borne diseases in both humans and animals (Freitas et al., 1978; Dumler and Bakken, 1998). In Brazil domestic dogs suffer heavy infestations of R. sanguineus and these ticks are a constant nuisance to the owners. The dog is considered to be the natural * To whom correspondence should be addressed at: Tel: 55 (016) 323–2500; fax: 55 (016) 322–4275; e-mail: [email protected]

916 host of R. sanguineus and does not develop resistance even after repeated feeding (Chabaud, 1950; Szabó et al., 1995b), whereas guinea pigs develop strong resistance after one infestation (Szabó et al., 1995b). Intradermal inoculation of whole unfed adult R. sanguineus extract of tick sensitized hosts (Szabó et al., 1995a) induced an immediate and a delayed hypersensitivity reaction in guinea pigs. Dogs, on the other hand, developed a strong immediate reaction, but a second peak in skin reaction was not observed, suggesting that cellular immunity is an important mechanism of resistance to R. sanguineus ticks. These differences in the reaction between natural and unnatural hosts to the R. sanguineus may be better understood if the inflammatory reactions at tick feeding sites, particularly cell migration, are characterized. Preliminary data indicated that, towards the end of the feeding time (fast engorging period), neutrophils accumulate at feeding sites on naive or post-infested dogs (Theis and Budwiser, 1974; Szabó and Bechara, 1995). On the other hand, mononuclear cells on guinea pigs are predominant at the end of the feeding period (Szabó and Bechara, 1995). The objective of the present study was to compare the dynamics of inflammatory reaction, particularly cell migration, induced by R. sanguineus ticks on a natural nonresistant host (dog) to that induced on a resistant and unnatural host (guinea pig) during primary (naive hosts) and tertiary infestation.

Materials and methods Experimental animals: All animals used were tick-bite naive at the beginning of the experiments. Thirty-four month-old female albino guinea-pigs and five male and seven female five to six-month-old mongrel dogs were used in our experiments. Guinea pigs were supplied by São Paulo State University (Botucatu, SP) and dogs were born and raised in a tick-free pen at Jaboticabal, SP, Brazil. Ticks: engorged R. sanguineus females, originally collected from healthy dogs at the Veterinary College Hospital in Jaboticabal, Brazil, were used to establish a colony in the laboratory. The ticks were fed on naive guinea-pigs and kept under a constant temperature of 29 °C 6 1°, a relative humidity of 80% 6 5% and a 12:12 (L:D) photoperiod. Sensitization of hosts: guinea-pigs and dogs were infested twice with a 30-day interval between infestations. Each guinea pig was infested twice with four female and five male ticks and each dog with eight females and 10 males (primary infestation) and 25 females and 30 males (secondary infestation). Unfed ticks were placed inside feeding chambers as decribed by Bechara et al. (1994). Briefly, plastic tubes, with a diameter of 2.5 cm and 3 cm height, were glued with a non-hazardous preparation (Britannia Adhesives, UK) onto the shaved backs of the guinea pigs.

917 Tubes placed on dogs were 5 cm in diameter and Elizabethan collars were used to prevent grooming. Skin sample collections: biopsies were taken from feeding sites on hosts approximately 4, 24, 48 and 96 hours following tick attachment during primary or tertiary infestation. Samples from each time interval were obtained from at least three different animals. Biopsies were performed with the aid of a 5 mm diameter punch. Control skin samples were obtained from the centre of feeding chambers kept for 96 hours on non-parasitized guinea-pigs and dogs. Skin samples were immediately immersed in fixative (buffered formalin, pH 7.0). Other fixatives such as BouinDubocsq or unbuffered formalin were also tested but did not allow the proper staining and distinction of basophils and eosinophils. Histotechnological processing: skin samples were kept for 24 hours in the fixative, embedded in paraffin and processed according to routine histological techniques. Each biopsy was serially sectioned at a thickness of 4 mm, and stained with Hematoxilin-eosin and May Grünwald-Giemsa. Section analysis: sections were categorized into four zones according to distance from tick-attachment sites (Figure 1) for analysis performed under light microscopy. Only sections displaying zone 1 and 2 were used for analysis, since these two zones include the central tick attachment lesion. General features were evaluated on Hematoxilin-eosin stained sections. Total cell counts were made on sections stained by May-Grünwald & Giemsa. For this purpose, cells from four areas of 0.0052 mm2 of the dermis were counted, two immediately below the epidermis and the cement cone (A-area) and another two 1 mm deep in the dermis (B-area). Means of each area were used for further analysis. The counting area was delimited by a Reichert integrating graticule (Austria/PK 6,3 3 mn) on oil immersion fields (objective 1003). Differential cell counts were performed on the same sections and areas used for total cell counts. In the presence of a feeding cavity (area of liquefative necrosis in the dermis of the host which sometimes occurred under the tick attachment site) cells surrounding this cavity were counted. Ultrastructural identification of basophils in the skin of guinea pigs exposed to three infestations with Rhipicephalus sanguineus adults: it is very difficult to identify basophils under light microscopy. In order to validate our observations under light microscopy, an ultrastructural analysis was also performed. For this purpose skin fragments from a 24-hour tick attachment site, during third infestation of a guinea pig, were fixed in 2.0% buffered glutaraldehyde. Routine double fixation and embedding in Araldite 502 were used (Luft, 1961). Thin sections were obtained with the aid of a Sorvall 5000 ultramicrotome, stained with uranyl acetate and lead citrate and examined in a Phillips ME-201 electron microscope. Statistical analysis: the statistical significance of the differences in cell counts following primary and tertiary infestations and between host species were determined using Mann–Whitney’s U test and between A and B areas using Wilcoxon’s matched pairs test (P , 0.05). Statistical differences among time intervals within host species

918

Zone 1: presence of an epidermal rupture matching the central channel of the tick’s attachment cement cone; zone 2: presence of a big, bell-shaped, mass of cement cone on the skin’s surface, but with lack of epidermal rupture; zone 3: presence of a flattened mass of cement cone on the epidermis suggesting surroundings of tick’s attachment site; zone 4: absence of cement cone. Figure 1. Diagram of the Rhipicephalus sanguineus tick attachment site at the skin of a host.

and infestations were determined by Kruskal–Wallis test. In case of significance, means were compared by Mann–Whitney’s U test (P , 0.05).

Results General features Dogs did not appear to be irritated by tick infestation and local macroscopic alterations were limited to skin thickening in areas where many ticks completed feeding close to each other. Guinea pigs, on the other hand, displayed a strong local reaction to ticks, mainly during tertiary infestations. Intense hyperemia, swelling, fluid exudation and even necrosis could be seen. Some microscopic features were common in many observed sections, irrespective of the host, infestation or time interval. Ticks, or parts of them, could be seen with

919 their hypostomes embedded in a cement cone. Tick attachment was always superficial being restricted to the epidermis and occasionally upper layers of the dermis. Cement cone covered a small area of the epidermis (0.82 6 0.4 mm; n 5 35) and could be seen as an eosinophilic homogeneous mass widening progressively toward the hypostome. The cement was generally confined to the surface of the epidermis extending under the stratum corneum at its far ends. At tick attachment sites the epidermis was usually thick due to hyperplasia, spongiosis and intracellular oedema. Exocytosis was also often observed. A feeding cavity (vesicle filled with dead cells and debris) was sometimes present below the tick’s attachment site. Main dermal changes included a varying degree of cellular infiltration, oedema, micro-abscesses and occasionally, haemorrhage and necrosis. All the changes described above were confined to the immediate areas surrounding the tick attachment sites (zones 1–3). These were slight during first infestation, especially in dogs, but very pronounced during the third infestation, particularly in guinea pigs. Initial changes (4 hours) were minor. Cellular infiltration was diffuse, less intense and restricted to the superficial dermis below the cement cone. A more intense cellular infiltration could be seen 24 and 48 hours after infestation under the cement cone which reached the deeper layers of the dermis. At the same time, a deep perivascular dermatitis, discontinuous with the superficial dermal one, was seen at the dermal-hypodermal border. Up to this time (48 h) local inflammation displayed mainly exudative features. The most intense changes were seen after 96 hours when cellular infiltration from the superficial and deep dermis tended to meet, originating a widely-affected area with proliferative inflammatory features, particularly in the tertiary infestation of guinea pigs. By 96 hours the high number of neutrophils in third infestation of dogs suggested an acute supurative process. In guinea pigs and dogs, tertiary infestation was characterized by a wider and more intense infiltration. The highest cellular concentration at tick feeding sites was seen in dogs during the tertiary infestation (A-area by 96 h), but the affected area was usually larger and more deeply extended in guinea pigs, reaching as far as the hypodermis and zone 4. Cell counts Cell counts in the A-area are presented in Tables 1 to 4 and Figure 2. Cell population was less dense and more heterogeneous at B-area (1 mm deep in dermis) (data not shown). These results indicate that A-area, displaying a higher concentration of specific cell populations, represents the main chemotactic site induced by tick lesion and antigens. For this reason, unless otherwise stated, analysis was based on data from A-area only. Total cell concentration was higher during tertiary infestation on both hosts. Total cell counts also showed that beneath the cement cone there was a tendency with time towards increasing concentration of migrating cells, but during

920 Table 1. Total and differential cell counts (total and relative values) at different time intervals at the attachment sites of Rhipicephalus sanguineus adults on dogs during primary infestations. Results are expressed as means 6 standard deviation per mm2. Time interval

Total counts

Neutrophil

Eosinophil

Basophil

Mast cell

Mononuclear cells

sham* (n 5 3)

641a 6445

59 6 64a (9%)

0a (0%)

0a (0%)

54 6 56a (8%)

528 6 432a (82%)

4h (n 5 3)

5449bc 65031

3469 6 4628a (64%)

0a (0%)

0a (0%)

139 6 85a (3%)

1809 6 1183a (33%)

24 h (n 5 4)

5048abc 66256

2757 6 4998a (55%)

780 6 1500a (15%)

0a (0%)

33 6 44a (1%)

1476 6 435a (29%)

48 h (n 5 5)

2192ab 61589

574 6 865a (26%)

65 6 144a (3%)

0a (0%)

205 6 269a (9%)

1348 6 1150a (61%)

96 h (n 5 6)

7564c 63446

6818 6 4309a (90%)

0a (0%)

0a (0%)

131 6 320a (2%)

615 6 774a (8%)

* Cell population in a non-infested healthy skin sample; n 5 number of samples. Means in a column followed by the same letter do not differ significantly.

Table 2. Total and differential cell counts (total and relative values) at different time intervals at the attachment sites of Rhipicephalus sanguineus adults on dogs during tertiary infestation. Results are expressed as means 6 standard deviation per mm2. Time interval

Total counts

Neutrophil

Eosinophil

Basophil

Mast cell

Mononuclear cells

sham* (n 5 3)

641a 6445

59 6 64a (9%)

0a (0%)

0a (0%)

54 6 56a (8%)

528 6 432a (82%)

4h (n 5 4)

2932ab 62638

1116 6 1723ab (38%)

26 6 52a (1%)

0a (0%)

102 6 96a (3%)

1688 6 1971a (58%)

24 h (n 5 4)

7067bcd 66418

5766 6 6445bc (82%)

0a (0%)

0a (0%)

155 6 219a (2%)

1146 6 446a (16%)

48 h (n 5 4)

8365c 61835

5995 6 3337b (72%)

148 6 252a (2%)

0a (0%)

164 6 153a (2%)

2057 6 1415a (25%)

96 h (n 5 6)

14712d 64746

13350 6 4515c (91%)

0a (0%)

0a (0%)

16 6 38a (0.1%)

1335 6 775a (9%)

* Cell population in a non-infested healthy skin sample; n 5 number of samples. Means in a column followed by the same letter do not differ significantly.

primary infestation a drop in cell counts was observed within 24 and 48 hours following tick attachment in both host species. Differential cell counts displayed sharp differences between the two host species. Dogs reacted overwhelmingly with neutrophils, especially at the end of the feeding periods, and also with mononuclear and mast cells. Neutrophils concentrated close to

921 Table 3. Total and differential cell counts (total and relative values) at different time intervals at the attachment sites of Rhipicephalus sanguineus adults on guinea pigs during primary infestations. Results are expressed as means 6 standard deviation per mm2. Time interval

Total counts

Neutrophil

Eosinophil

Basophil

Mast cell

Mononuclear cells

sham* (n 5 5)

1859a 6616

35 6 39a (2%)

172 6 263a (9%)

0a (0%)

35 6 63a (2%)

1617 6 564a (87%)

4h (n 5 3)

6410bd 62749

170 6 162a (3%)

343 6 293a (5%)

2898 6 4125b (45%)

142 6 181a (2%)

2856 6 1692a (45%)

24 h (n 5 4)

3461abc 61413

11 6 23a (0.3%)

248 6 306a (7%)

314 6 234b (9%)

72 6 60a (2%)

2851 6 1141a (82%)

48 h (n 5 4)

2548ac 6921

64 6 128a (3%)

599 6 397a (24%)

109 6 179a (4%)

60 6 75a (2%)

1715 6 775a (67%)

96 h (n 5 4)

9663d 61694

287 6 216a (3%)

3667 6 858a (38%)

1739 6 2299b (18%)

25 6 50a (0.3%)

3817 6 1082a (40%)

* Cell population in a non-infested skin sample; n 5 number of samples. Means in a column followed by the same letter do not differ significantly.

Table 4. Total and differential cell counts (total and relative values) at different time intervals at the attachment sites of Rhipicephalus sanguineus adults on guinea pigs during tertiary infestations. Results are expressed as means 6 standard deviation per mm2. Time interval

Total counts

Neutrophil

Eosinophil

Basophil

Mast cell

Mononuclear cells

sham* (n 5 5)

1859a 6616

35 6 39a (2%)

172 6 263a (9%)

0a (0%)

35 6 63a (2%)

1617 6 564a (87%)

4h (n 5 3)

4679b 6867

0a (0%)

275 6 254ab (6%)

2104 6 1291b (45%)

0a (0%)

2300 6 923a (49%)

24 h (n 5 4)

9231c 62813

22 6 45a (0.2%)

464 6 385ab (5%)

6744 6 3540b (73%)

118 6 181a (1%)

1838 6 537a (20%)

48 h (n 5 4)

9772c 61494

43 6 50a (1%)

2706 6 3397bc (28%)

4531 6 3248b (46%)

59 6 118a (1%)

2373 6 1069a (24%)

96 h (n 5 4)

9856c 63455

56 6 69a (1%)

3092 6 2111c (31%)

4699 6 2468b (48%)

63 6 127a (1%)

1944 6 1078a (20%)

* Cell population in a non infested skin sample; n 5 number of samples. Means in a column followed by the same letter do not differ significantly.

the attachment sites and mononuclear cells were seen more peripherally and the range of each cell type varied according to infestation and time interval. Many mast cells were seen surrounding blood vessels, but mostly away from the counting area. Guinea pigs, on the other hand, reacted with a variable amount of mononuclear cells, basophils and eosinophils. These granulocytes outnumbered other cell types during

922

dg 5 dog; gp 5 guinea pig; p 5 primary infestation; t 5 tertiary infestation. Figure 2. Total and differential cell counts at Rhipicephalus sanguineus tick attachment sites at different time intervals of feeding during primary and tertiary infestations of dogs and guinea pigs.

tertiary infestations of guinea pigs. Basophils were identified by their typical nuclei and large cytoplasmic granules staining purple in May-Grünwald Giemsa (Figure 3). In contrast, eosinophil granules, stained purplish red with the same stain (Figure 3). These two cell types looked alike in Hematoxylin-eosin sections where their granules stained red. Ultrastructural analysis of the infested skin of guinea pigs confirmed the identity of eosinophils and basophils. Eosinophils were seen with coarsely reticular zones and sharply demarcated electron dense cores, whereas the majority of basophil granules had a parallel banding or a honeycomb pattern. The normal skin of non-infested dogs and guinea pigs displayed a small mast cell and mononuclear cell population and no changes were observed. Statistical analysis All cell counts were characterized by high standard deviations but many significant differences were observed. B-area (data not shown) differed significantly from Aarea in total and differential cell counts. Dogs presented significantly higher total and neutrophil cell counts at the dermal-epidermal border (A-area) both in naive and

923

Figure 3. Photomicrograph of Rhipicephalus sanguineus tick attachment site after 96 hours on a guinea pig during third infestation. Notice eosinophils (e), basophils (b) and mononuclear cells (m). MayGrünwald Giemsa (objective 1003).

infested animals. Moreover, the percentage of neutrophils was also higher close to the attachment site (A-area). Guinea pigs had significantly higher total, basophil, eosinophil and monocyte cell counts at dermal-epidermal border (A-area) in naive and infested animals. It must be stressed, however, that monocytes, although in higher total numbers in A-area, were a lower percentage in relation to that seen in Barea. Results shown above refer to A-area only. Infestations of dogs and guinea pigs were compared. Control skin samples (noninfested) of guinea pigs displayed significantly higher total and mononuclear cell

924 Table 5. Significance between differential cell counts of dogs and guinea pigs at the R. sanguineus feeding site at different time intervals after infestation. Infestation

4h

24 h

48 h

96 h

Primary

basophil

basophil mononuclear

eosinophil

neutrophil basophil eosinophil mononuclear

Tertiary

neutrophil basophil

neutrophil basophil eosinophil



counts in relation to dogs. Total and mast cell counts at tick attachment sites did not differ significantly between these host species. Significance between differential cell counts are presented in Table 5. Primary and tertiary infestations were also compared. Tertiary infestations of dogs presented significantly higher total and neutrophil cell counts (48 and 96 hours of infestation) when compared to primary infestations. At the same time, guinea pigs presented higher total and basophil cell counts during tertiary infestations (24 and 48 hours of tick attachment). Significance of differences among time intervals within host species are presented in Tables 1–4.

Discussion Our results show that the difference in resistance between dogs and guinea pigs to R. sanguineus is accompanied by differences in cellular infiltration at tick attachment sites. The rationale behind this approach is to compare reactions of a stronglyresistant host with one that has no resistance, and thus learn about effective and inadequate immune reactive mechanisms to ticks. Alterations such as thickened epidermis due to hyperplasia, varying degrees of epidermal inter- and intracellular oedema, dermal infiltration, oedema haemorrhage and necrosis are common alterations induced by many noxious stimuli and might be regarded as non specific. Such changes agree with earlier observations on tick attachment sites (Walker and Fletcher, 1986). Cell counts were characterized by high standard deviations which indicate elevated individual variation. A-area displayed the highest concentration of specific cell populations and thus represented the main chemotactic site induced by tick lesion and antigens. For this reason analysis was based mainly on data from the A-area. Total cell counts, composed overwhelmingly of granulocytes (basophils and eosinophils in guinea pigs and neutrophils in dogs), showed significant increase from primary to tertiary infestations indicating the role for immune modulation during

925 tertiary infestation in both hosts. Primary infestation was characterised by a drop in total cell numbers, mainly granulocytes, from 4 to 24 and 48 hours of feeding. This drop might have occurred due to a decrease in cell migration and/or a local increase in cell death rate but reasons for this are uncertain. Cells seemed to infiltrate a larger area in guinea pigs, but 96 hours after tertiary infestation a higher cell density (40%) was observed in dogs at A-area, when compared with guinea pigs. This intense reaction in dogs seems not to harm ticks as this host does not develop resistance to R. sanguineus (Chabaud, 1950; Szabó et al., 1995b). The main difference between the two host species at tick attachment sites was observed for differential cell counts. It should be stressed, however, that the methodology used was unable to distinguish among mononuclear cells or point out activation state of the granulocytes, which are very important in defining resistance mechanisms. Mononuclear cell counts thus included macrophages, lymphocytes and fibroblasts as it was impossible to accurately distinguish these cell types under light microscopy. Tertiary infestation in guinea pigs elicited an increase in eosinophils and basophils. These cells have previously been observed in high numbers at feeding sites of ticks on different host species (reviewed by Wikel, 1996), but not on dogs. Nevertheless, hypersensitive dogs react with cutaneous basophilia against fleas (Halliwell and Schemmer, 1987), indicating that this host species has the potential to react with this granulocyte. Migration or activation of basophils might be suppressed in dogs infested by R. sanguineus ticks. During primary infestation guinea pigs reacted with a higher percentage of mononuclear cells but already had significant numbers of basophils and eosinophils. This suggests an innate reactive pattern of guinea pigs to ticks which was enhanced by sensitization. This reaction might already be present during first infestation as shown by the lower egg mass weights and longer engorging periods of female R. sanguineus ticks fed on tick-bite naive guinea pigs when compared to those fed on naive dogs (Szabó et al., 1995b). The importance of basophils and eosinophils in resistance against ticks is suggested by the decrease of guinea pig resistance following treatment of sensitized animals with anti-basophil and anti-eosinophil serum (Brown et al., 1982). These experiments also showed that anti-basophil serum impairs resistance more efficiently. Not much is known about the mechanisms by which this granulocyte interferes with tick feeding. Some authors believe that basophils participate at local anaphylactic reactions by releasing mediators which could harm tick feeding and salivation (Paine et al., 1983). Histamine, produced by basophils and mast cells induces a strong vascular permeability leading to oedema (Cotran et al., 1994). According to Ribeiro (1989), oedema decreases nutritive element disposal at tick attachment sites. The impairing action of eosinophils is also not known. Brown et al. (1982) suggested an involvement of eosinophil granule protein peroxidase in the expression

926 of resistance against ticks. Similar proteins are known to be able to impair nematodes (Scepek et al., 1994). The presence of eosinophils could also be explained by their capacity to inactivate histamine (Zucker-Franklin, 1990). Lymphocytes and macrophages are probably involved in the establishment and control of the immune reaction and take part in the late inflammatory process. Cutaneous reactions of dogs observed in our work were characterized by a massive neutrophil infiltration, regardless of the number of infestations and the time of attachment and confirmed previous observations (Theis and Budwiser, 1974). This infiltration could be mediated, at least partially, by mast cells known to produce Interleukin-8, a strong chemotactic factor for neutrophils (Moller et al., 1993). As observed from the results, dogs have many mast cells surrounding blood vessels and react with a strong immediate hypersensitivity reaction during cutaneous hypersensitivity tests with whole unfed adult R. sanguineus extract (Szabó et al., 1995a) which also suggests a mast cell involvement. It is known, however, that dogs have a high neutrophil/mononuclear cell ratio in the blood and a greater tendency towards involvement of neutrophils in the haematological response to a variety of stimuli (Jain, 1986). Moreover Galkowska and Olszewski (1986) showed that large numbers of neutrophils are present in the lymphatic system of dogs, but not in other species. It has also been shown (Comer, 1988) that acute spontaneous lesions in inflammatory canine skin diseases, such as atopic dermatitis and allergic contact dermatitis, are characterized by pruritus and a cellular infiltration rich in neutrophils, erythema and edema. These observations suggest that a neutrophilic reaction in dogs might be, at least partially, a host-linked pattern. Neutrophils can potentially harm tissues by means of lytic enzymes and reactive oxygen species (Weiss, 1989). The lack of strong reactions in our observations suggests that R. sanguineus is able to manipulate the local inflammatory reactions of the dog, because a potentially harmful neutrophilic response does not affect ticks, and causes no significant damage to the host’s skin, with no pain or pruritus. Thus, an important host resistance behaviour pattern – the self grooming – was not seen. Ribeiro and co-workers (1990) observed that the saliva from another dog tick, Ixodes dammini, can inhibit the aggregation, enzyme and free radical secretion and phagocytosis of rat neutrophils. Inokuma and co-workers (1997) observed that R. sanguineus tick infestation suppressed the neutrophil chemilluminescence stimulated by opsonized zymosan of circulating dog cells. It may be supposed that R. sanguineus tick saliva also has a dog neutrophil deactivating capacity. Neutrophils are known to have, apart from their phagocytic activity, a limited secretory capacity with immuno-modulating properties (Lloyd and Oppenheim, 1992). Regarding this matter, Galkowska and co-workers (1989) demonstrated that canine neutrophils spontaneously release a factor that suppresses lymphocyte proliferation responses. It is also known that neutrophils highly express calcium-binding protein MRP-14 which deactivates activated macrophages in vitro (Giorgi et al.,

927 1998). One could speculate that dog neutrophils at R. sanguineus tick attachment sites suppress inflammatory activity. In conclusion these results depict the huge differences in histological reaction of dogs and guinea pigs to infestation by R. sanguineus tick. These differences appear to be linked to the absence of resistance in dogs to R. sanguineus, but many features remain to be elucidated. A better characterization of mononuclear cells at tick feeding sites is highly desirable to the understanding of ongoing immune processes. The role of basophils in parasitic diseases remains speculative and much remains to be studied about this granulocyte.

Acknowledgements We thank Mrs Maria Inês Yamazaki de Campos, Mrs Shirley Meire da Silva Silva and Miss Francisca de Assis Ardisson for technical assistance; Dr Idércio Luiz Sinhorini and Dr Laurelúcia de Orives Lunardi for help with electronmicroscopy; Dr Genner Tadeu Pereira for statistical assistance; Sérgio Fonseca, Purina Nutrimentos, Brazil, for providing commercial dog food; and Fundacção de Amparo a` Pesquisa do Estado de São Paulo (FAPESP) and Conselho Nacional de Desenvolvimento Cient´ıfico e Tecnológico (CNPQ) for grants. We also thank Dr Laura Fielden, Berry College, Georgia, USA, for reviewing the manuscript. The results of this study represent part of a PhD thesis by M.P.J. Szabó, at the Department of Pathology, Faculdade de Medicina Veterinária e Zootecnia da USP, São Paulo, Brasil.

References Bechara, G.H., Szabó, M.P.J., Mukai, L.S. and Rosa, P.C.S. 1994. Immunisation of dogs, hamsters and guinea pigs against Rhipicephalus sanguineus using crude unreleased adult tick extracts. Vet. Parasitol. 52: 79–90. Brown, S.J., Galli, J.S., Gleich, G.J. and Askenase, P.W. 1982. Ablation of immunity to Amblyomma americanum by anti-basophil serum: cooperation between basophils and eosinophils in expression of immunity to ectoparasites (ticks) in guinea-pigs. J. Immunol. 129 (2): 790–796. Chabaud, A.G. 1950. L’infestation par des ixodinés provoque-t-elle une immunité chez l-hôte (2me note). Ann. Parasitol. 25 (5–6): 474–479. Comer, K.M. 1988. Carpet deodorizer as a contact allergen in a dog. J.A.V.M.A. 193: 1553–1554. Cotran, R.S., Kumar, V. and Robbins, S.L. 1994. Inflammation and Repair. In: Robbins Pathologic Basis of Disease, R.S. Cotran, V. Kumar and S.L. Robbins (eds). W.B. Saunders Company, 5th ed. Dumler, J.S. and Bakken, J.S. 1998. Human ehrlichioses: newly recognized infections transmitted by ticks. Annual Ver. Med. 49: 201–213. Fielden, L.J., Rechav, Y. and Bryson, N.R. 1992. Acquired immunity to larvae of Amblyomma marmoreum and A. hebraeum by tortoises, guinea-pigs and guinea-fowl. Med. Vet. Entomol. 6: 251–254. Freitas, M.G., Costa, H.M.A., Costa, J.O. and Iide, P. 1978. Entomologia e Acarologia Médica e Veterinária. Editora Rabelo e Brasil, Belo Horizonte, Brasil. Galkowska, H. and Olszewski, W.L. 1986. Cellular composition of lymph in experimental lymphedema. Lymphology 19: 139–145.

928 Galkowska, H., Dabrowski, M. and Olszewski, W.L. 1989. Spontaneously active suppressive cells in the canine peripheral blood. Vet. Immunol. Immunopathol. 20: 101–108. Giorgi, R., Pagano, R.L., Dias, M.A., Aguiar-Passeti, T., Sorg, C. and Mariano, M. 1998. Antinociceptive effect of the calcium-binding protein MRP-14 and the role played by neutrophils on the control of inflammatory pain. J. Leukoc. Biol. 64(2): 214–220. Halliwell, R.E.W. and Schemmer, R.K. 1987 The role of basophils in the immunopathogenesis of hypersensitivity to fleas (Ctenocephalides felis) in dogs. Vet. Immunol. Immunopathol.15: 203–13. Inokuma, H., Hara, Y., Aita, T. and Onishi, T. 1997. Effect of infestation with Rhipicephalus sanguineus on neutrophil function in dogs. Med. Vet. Entomol. 11: 401–403. Jain, N.C. 1986. The neutrophils. In: Schalm’s Veterinary Hematology, 4th Edn, pp. 676–730. Lea and Feibiger Press, Philadelphia. Lloyd, A.R. and Oppenheim, J.J. 1992. Poly’s lament: the neglected role of the polymorphonuclear neutrophil in the afferent limb of the immune response. Immunol. Today 13(5): 169–172. Luft, J.H. 1961. Improvements in epoxi resin embedding methods. J. Biochem. Cytol. 9: 409–417. Moller, A., Lippert, U., Lessmann, D., Kolde, G., Hamann, K., Welker, P., Schadendorf, D., Rosenbach, T., Luger, T. and Czarnetzki, B.M. 1993. Human mast cells produce Il-8. J. Immunol. 151(6): 3261–3266. Paine, S.H., Kemp, D.H. and Allen, J.R. 1983. In vitro feeding of Dermacentor andersoni (Stiles): effects of histamine and other mediators. Parasitology 86: 419–428. Pegram, R.G., Clifford, C.M., Walker, J.B. and Keirans, J.E. 1987. Clarification of the Rhipicephalus sanguineus group (Acari, Ixodoidea, Ixodidae). I. R. sulcatus (Neuman, 1908) and R. turanicus (Pomerantsev, 1936). Sys. Parasitol 10: 3–26. Randolph, S.E. 1979. Population regulation in ticks: the role of acquired resistance in natural and unnatural hosts. Parasitology 79: 141–156. Rechav, Y. 1992. Naturally acquired resistance to ticks – a global view. Insect. Sci. Applic. 13(4): 405–504. Ribeiro, J.M.C. 1989. Role of saliva in tick/host interactions. Exp. Appl. Acarol. 7: 15–20. Ribeiro, J.M.C., Weiss, J.J. and Telford III, S.R. 1990. Saliva of the tick Ixodes dammini inhibits neutrophil function. Exp. Parasitol. 70: 382–8. Scepek, S., Moqbel, R. and Lindau, M. 1994. Compound exocytosis and cumulative degranulation by eosinophils and their role in parasite killing. Parasitol. Today 10(7): 276–278. Szabó, M.P.J. and Bechara, G.H. 1995. An insight into the histopathology caused by the tick Rhipicephalus sanguineus (Acarina: Ixodidae) in the skin of previously infested, vaccinated or tickbite naive dogs, guinea pigs and hamsters. Braz. J. vet. Res. anim. Sci. 32(1): 37–42. Szabó, M.P.J., Morelli Jr., J. and Bechara, G.H. 1995a. Cutaneous hypersensitivity induced in dogs and guinea pigs by extracts of the tick Rhipicephalus sanguineus (Acari: Ixodidae). Exp. Appl. Acarol. 19: 723–730. Szabó, M.P.J., Mukai, L.S., Rosa, P.C.S. and Bechara, G.H. 1995b. Differences in the acquired resistance of dogs, hamsters, and guinea pigs to repeated infestations with adult ticks Rhipicephalus sanguineus (Acari: Ixodidae). Braz. J. vet. Res. anim. Sci. 32(1): 43–50. Theis, J.H. and Budwiser, P.D. 1974. Rhipicephalus sanguineus: sequential histopathology at the hostarthropode interface. Exp. Parasitol. 36: 77–105. Walker, A.R. and Fletcher, J.D. 1986. Histological study of the attachment sites of adult Rhipicephalus appendiculatuson rabbits and cattle. Int. J. Parasitol. 16: 399–413. Wikel, S.K. 1996. Immunology of the tick-host interface. In: The immunology of Host-ectoparasitic arthropod relationships, S.K. Wikel (ed.), pp. 204–231. Cab International, UK. pp. 204–231. Weiss, S.J. 1989. Tissue destruction by neutrophils. N. Engl. J. Med. 320(6): 365–376. Zucker-Franklin, D. 1990. Eosinophils: Morphology, production, chemistry and function. In: Hematology, W.J. Williams, E. Beutler, A.J. Erslev and M.A. Lichtman, (eds), 4th edn, pp. 835–855. McGrawHill, Inc.