Experimental Infection of Rhipicephalus sanguineus Ticks with the Bacterium Rickettsia rickettsii,Using Experimentally Infected Dogs

Abstract

We evaluated if Rickettsia rickettsii-experimentally infected dogs could serve as amplifier hosts for Rhipicephalus sanguineus ticks. In addition, we checked if Rh. sanguineus ticks that acquired Ri. rickettsii from dogs could transmit the bacterium to susceptible hosts (vector competence), and if these ticks could maintain the bacterium by transstadial and transovarial transmissions. Uninfected larvae, nymphs, and adults of Rh. sanguineus were allowed to feed upon three groups of dogs: groups 1 (G1) and 2 (G2) composed of Ri. rickettsii-infected dogs, infected intraperitoneally and via tick bites, respectively, and group 3 composed of uninfected dogs. After larval and nymphal feeding on rickettsemic dogs, 7.1–15.2% and 35.8–37.9% of the molted nymphs and adults, respectively, were shown by polymerase chain reaction (PCR) to be infected by Ri. rickettsii, confirming that both G1 and G2 dogs were efficient sources of rickettsial infection (amplifier host), resulting in transstadial transmission of the agent. These infected nymphs and adults successfully transmitted Ri. rickettsii to guinea pigs, confirming vector competence after acquisition of the infection from rickettsemic dogs. Transovarial transmission of Ri. rickettsii was observed in engorged females that had been infected as nymphs by feeding on both G1 and G2 dogs, but not in engorged females that acquired the infection during adult feeding on these same dogs. In the first case, filial infection rates were generally <50%. No tick exposed to G3 dogs was infected by rickettsiae in this study. No substantial mortality difference was observed between Ri. rickettsii-infected tick groups (G1 and G2) and uninfected tick group (G3). Our results indicate that dogs can be amplifier hosts of Ri. rickettsii for Rh. sanguineus, although only a minority of immature ticks (<45%) should become infected. It appears that Rh. sanguineus, in the absence of horizontal transmission, would not maintain Ri. rickettsii through successive generations, possibly because of low filial infection rates.

Introduction

T he bacterium Rickettsia rickettsii is the etiological agent of Rocky Mountain spotted fever (RMSF), an acute febrile disease that has been called in Brazil as Brazilian spotted fever. During the last two decades, there has been a clear reemergence of RMSF in southeastern Brazil, where nearly 350 laboratory-confirmed cases (case fatality ∼30%) were reported (Labruna 2009). RMSF has been reported in America, from Canada to Argentina, where it is vectored by different tick species. In the United States, the tick species Dermacentor andersoni and Dermacentor variabilis are the main vectors through the western and eastern parts of the country, respectively, whereas Rhipicephalus sanguineus has been implicated as vector in a few particular areas (Demma et al. 2005, Dumler and Walker 2005). In Mexico, Rh. sanguineus has been considered the main vector, together with Amblyomma cajennense in some areas (Bustamante and Varela 1947, Barrera 1956). In central and South America, A. cajennense is the main vector, except for the São Paulo city metropolitan area, where Amblyomma aureolatum is the most important vector (Labruna 2009). Recently, there have been independent reports of Rh. sanguineus naturally infected by Ri. rickettsii in three RMSF endemic areas in southeastern Brazil, suggesting that Rh. sanguineus could play a role in the epidemiology of RMSF in this country (Cunha et al. 2009, Gehrke et al. 2009, Moraes-Filho et al. 2009). In all these cases, infected ticks were collected on dogs, the primary host of Rh. sanguineus. Although widespread on dogs in Latin America, Rh. sanguineus has been only occasionally associated with human infestations (Guglielmone et al. 2006).

Experimental studies in the United States demonstrated that Ri. rickettsii is transmitted transovarially and transstadially in a number of tick species, including Rh. sanguineus (Parker et al. 1933). However, under natural conditions, tick infection rates by Ri. rickettsii are very low (usually <1%), mainly because it has been shown that Ri. rickettsii is pathogenic for ticks (Burgdorfer 1988, Niebylski et al. 1999). Thus, the role of amplifier hosts in the ecology of RMSF is crucial. Amplifier hosts are vertebrate animals that develop rickettsemia for some days, when new uninfected ticks become infected and start new lineages of infected ticks within the tick population (Burgdorfer 1988, Dumler and Walker 2005). Recent experimental studies in Brazil showed that both opossums (Didelphis aurita) and capybaras (Hydrochoerus hydrochaeris) are competent amplifier hosts of Ri. rickettsii for A. cajennense ticks (Horta et al. 2009, Souza et al. 2009). As these two wild vertebrate species are abundant in many RMSF endemic areas in Brazil, where they usually carry high A. cajennense burdens, they have been implicated to play a significant role in the ecology of Ri. rickettsii (Labruna 2009). The domestic dog is also abundant in many RMSF endemic areas of Brazil, where besides Rh. sanguineus, dogs are also infested by A. cajennense and/or A. aureolatum ticks (Cunha et al. 2009, Moraes-Filho et al. 2009). Thus, our study evaluated if dogs experimentally infected with a Brazilian strain of Ri. rickettsii (strain Taiaçu) could serve as amplifier hosts for Rh. sanguineus ticks. In addition, we checked if Rh. sanguineus ticks that acquired Ri. rickettsii from a dog could transmit the bacterium to a susceptible host (vector competence), as well as if they could maintain the bacterium by transstadial and transovarial transmissions.

Materials and Methods

Dogs as experimental amplifier hosts of Ri. rickettsii for Rh. sanguineus ticks

This part of the study tested if Ri. rickettsii-experimentally infected dogs could be an efficient source of Ri. rickettsii infection (i.e., amplifiers hosts) for Rh. sanguineus ticks. For this purpose, we used eight, 6-month-old female dogs seronegative for rickettsiosis and ehrlichiosis, as previously reported elsewhere (Piranda et al. 2008). These dogs were provided by an animal room where they were reared free of natural infestations by ectoparasites, which were absent also during our study. Dogs were assigned to three experimental groups: group 1 (G1) was composed of three dogs that were each inoculated intraperitoneally with ∼3000 Vero cells 100% infected by Ri. rickettsii strain Taiaçu; group 2 (G2) was composed of three dogs that were each infested with five pairs of A. aureolatum adult ticks 100% infected by Ri. rickettsii, derived from an experimentally infected colony in our laboratory at the University of São Paulo; and group 3 (G3) was composed of two dogs that were left uninfected as the control group, which were each infested with five pairs of uninfected A. aureolatum adult ticks, derived from our laboratory colony, as previously reported (Piranda et al. 2008). Two days postinoculation or infestation (DPI), each dog was infested with 200 larvae, 600 nymphs, and 20 adult couples, ∼25 days old, of Rh. sanguineus derived from the fifth laboratory generation of a pathogen-free colony from the Department of Parasitology of the Institute of Veterinary, Federal Rural University of Rio de Janeiro. This colony originated from engorged females collected on naturally infested dogs in Seropédica (22°45′S, 43°41′W), state of Rio de Janeiro. These ticks were morphologically identified as Rh. sanguineus, according to Walker et al. (2003). In addition, a fragment of the 16S rDNA gene of a representative specimen was sequenced and showed to be 99.6% (443/445) identical to a Rh. sanguineus corresponding sequence available in GenBank (AY883872).

For tick infestations on each of the G1 dogs, a white cotton sleeve (15 × 15 cm) was glued to the shaved back as previously described (Piranda et al. 2008), and this sleeve received larvae, nymphs, and adults of Rh. sanguineus all together on DPI 2. For G2 and G3 dogs, two cotton sleeves were glued to the shaved back of each dog, one sleeve receiving A. aureolatum adult ticks on DPI 0, and the second sleeve receiving Rh. sanguineus larvae, nymphs, and adults on DPI 2. The sleeves were opened daily, and detached engorged ticks were removed, counted, and left in an incubator at 25°C and 85% relative humidity (RH) for molting (in the case of engorged larvae and nymphs) or egg laying (in the case of engorged females).

Random samples of unfed nymphs and adults that had molted from engorged ticks, and unfed larvae derived from engorged females of Rh. sanguineus were selected to be tested by PCR, when these ticks were ∼25 days old. In addition, the engorged females were also tested by PCR after they terminated oviposition in the incubator. For this purpose, ticks were submitted to DNA extraction by the guanidine thiocyanate method, as previously described (Labruna et al. 2008), and tested by PCR targeting a 401-bp fragment of the rickettsial gene gltA (Labruna et al. 2004). This DNA extraction protocol was shown to be 100% efficient for Ri. rickettsii-infected ticks (Labruna et al. 2008). Nymphs and adults were processed individually, whereas larvae were tested in pools of five larvae each. In each PCR run, three negative control tubes containing water were included, and also a positive control tube containing Rickettsia parkeri DNA.

Transstadial and transovarial transmission of Ri. rickettsii and vector competence of Rh. sanguineus

The maintenance of Ri. rickettsii through at least two consecutive developmental stages of Rh. sanguineus was evaluated, namely, from larvae to nymphs or from nymphs to adults (transstadial transmission), and from engorged females to larval offspring (transovarial transmission). Vector competence was measured as the ability of Ri. rickettsii-infected ticks to transmit the bacterium while feeding on guinea pigs under controlled conditions. For these purposes, we conducted three parallel experiments (I, II, and III). In experiment I, infestations started with unfed nymphs derived from the engorged larvae that had fed on G1, G2, and G3 dogs. These nymphs were allowed to feed on pathogen-free guinea pigs. Engorged nymphs were held for molting to adults in an incubator at 25°C and 85% RH. The resultant unfed adults were tested for rickettsial infection by PCR as described earlier.

In experiment II, infestations started with unfed adults derived from engorged nymphs that had fed on G1, G2, and G3 dogs. These adults were allowed to feed on pathogen-free guinea pigs. Engorged females were held for oviposition in an incubator at 25°C and 85% RH. The resultant unfed larvae were tested for rickettsial infection by PCR as described earlier.

In experiment III, infestations started with unfed larvae derived from engorged females that had fed on G1, G2, and G3 dogs. These larvae were allowed to feed on pathogen-free guinea pigs. Engorged larvae were held for molting in an incubator at 25°C and 85% RH. The resultant unfed nymphs were tested for rickettsial infection by PCR as described earlier.

Tick infestations on each guinea pig was performed inside a white cotton sleeve (8 × 8 cm) glued to the shaved back, as previously described (Labruna et al. 2008). Infested guinea pigs had their rectal temperature measured daily for 21 days, and then they were euthanized and their blood sera were tested for the presence of anti-Ri. rickettsii reactive antibodies by the immunofluorescence assay, as previously described (Horta et al. 2009). However, if a guinea pig presented fever (rectal temperature >40°C) during three consecutive days within this 21-day period, it was euthanized on the third febrile day for collection of blood and spleen. A spleen sample was submitted to DNA extraction using the DNeasy Tissue Kit (Qiagen, Chatsworth, CA) and then to the same PCR protocol described earlier. At the same time, a sample of fresh blood was inoculated into shell vials containing a monolayer of confluent Vero cells to isolate rickettsia, as previously described (Labruna et al. 2004). Rickettsial isolation was checked by Gimenez staining of scrapped cells (Labruna et al. 2004). If Rickettsia-like organisms were observed, cells were harvested and submitted to DNA extraction using the DNeasy Tissue Kit, followed by PCR targeting a portion of the rickettsial ompA gene (Regnery et al. 1991). Generated amplicons underwent DNA sequencing, and the resultant sequences were compared with GenBank data by Blast analysis (Altschul et al. 1990).

During the experiments, tick biological parameters were compared between Ri. rickettsii-infected and uninfected tick groups. For this purpose, feeding period, weight of engorged females and their corresponding egg masses, and egg mass% of hatching were compared by analysis of variance for random distributions and Mann–Whitney for nonrandom distributions. In addition, feeding, molting, and oviposition success were compared by the Fischer exact test or chi-square test (χ ²). Values were considered significantly different when p < 0.05. This study was previously approved by the Bioethical Committee in Animal Research of the Faculty of Veterinary Medicine of the University of São Paulo (protocol number 913/2006).

Results

Clinical and laboratory data of the three groups of dogs of this study have been presented elsewhere (Piranda et al. 2008). In summary, all G1 and G2 dogs developed signs of rickettsial infection, that is, fever, lethargy, anorexia, thrombocytopenia, anemia, and detectable levels of rickettsial DNA and Ri. rickettsii-reactive antibodies in their blood. In G1 dogs, rickettsemia started 2–4 DPI and lasted for 3–13 days. In G2 dogs, rickettsemia started 6–8 DPI and lasted for 3–7 days. In contrast, no clinical abnormalities, Rickettsia DNA, or Ri. rickettsii-reactive antibodies were detected in G3 dogs.

Dogs as experimental amplifier hosts of Ri. rickettsii for Rh. sanguineus ticks

Larvae, nymphs, and adults of Rh. sanguineus did feed on dogs for 4–11 days, coinciding with the rickettsemic period. PCR performed on random samples of unfed ticks, derived from engorged ticks that fed on rickettsemic dogs, showed that <45% of the larvae and nymphs acquired and maintained the rickettsial infection from the dogs to the subsequent tick developmental stage, nymphs and adults, respectively (Table 1). In all cases, larvae and nymphs exposed to G1 dogs (infected by Ri. rickettsii via intraperitoneal inoculation) resulted in higher proportions of Rickettsia-infected nymphs and adults, respectively, than ticks exposed to G2 dogs (infected by Ri. rickettsii via infected ticks).

Table 1.

Rickettsial Infection Determined by Polymerase Chain Reaction in Individual Rhipicephalus sanguineus Ticks After Exposition to Three Groups of Dogs

		No. of infected ticks/no. of tested ticks (% infection)
Dogs		Nymphs ^a	Adults ^b	Engorged females ^c	Larvae ^d
G1	1	4/24 (16.7)	10/26 (38.5)	4/4 (100)	0/60 (0)
	2	5/36 (13.9)	4/16 (25.0)	4/4 (100)	0/60 (0)
	3	8/52 (15.4)	19/45 (42.2)	4/4 (100)	0/60 (0)
	Total	17/122 (15.2)	33/87 (37.9)	12/12 (100)	0/180 (0)
G2	1	1/11 (9.1)	7/20 (35.0)	4/4 (100)	0/60 (0)
	2	4/63 (6.3)	15/41 (36.6)	4/4 (100)	0/60 (0)
	3	1/11 (9.1)	7/20 (35.0)	4/4 (100)	0/60 (0)
	Total	6/85 (7.1)	29/81 (35.8)	12/12 (100)	0/180 (0)
G3	1	0/16 (0)	0/22 (0)	0/3 (0)	0/30 (0)
	2	0/15 (0)	0/16 (0)	0/3 (0)	0/30 (0)
	Total	0/31 (0)	0/38 (0)	0/6 (0)	0/60 (0)

G1 consisted of three dogs inoculated intraperitoneally with Rickettsia rickettsii; G2 consisted of three dogs infested with Ri. rickettsii-infected Amblyomma aureolatum adult ticks; G3 consisted of two dogs infested with uninfected A. aureolatum adult ticks, being the uninfected control group.

Unfed nymphs that molted from larvae fed on dogs.

Unfed adults that molted from nymphs fed on dogs.

Engorged females fed on dogs, tested at the end of oviposition.

Unfed larvae derived from the offspring of engorged females fed on dogs. Larvae were tested in pools, each pool containing five individuals.

Random samples of Rh. sanguineus engorged females, which fed on either G1 or G2 rickettsemic dogs, showed to be 100% infected by rickettsia, as demonstrated by PCR performed on these females at the end of oviposition. However, none of the hatched larvae derived from these females were shown to contain rickettsial DNA by PCR (Table 1). None of the ticks (larvae, nymphs, and females) exposed to G3 dogs (uninfected control) resulted in PCR-positive ticks.

Transstadial and transovarial transmission of Ri. rickettsii and vector competence of Rh. sanguineus

After uninfected larvae were allowed to feed on G1 and G2 dogs, it was estimated by PCR that only 7.1–15.2% of the subsequent unfed nymphs were infected by rickettsiae. Part of these nymphs (experiment I) were allowed to feed on 12 susceptible guinea pigs, from which only one became infected by Ri. rickettsii, as demonstrated by high fever starting at 8 DPI, PCR-positive spleen, and isolation of rickettsia from blood by the shell vial technique (Table 2). The rickettsial isolate was confirmed to be Ri. rickettsii by PCR plus DNA-sequencing. After uninfected nymphs were allowed to feed on G1 and G2 dogs, it was estimated by PCR that 35.8–37.9% of the subsequent unfed adults were infected by rickettsiae. Part of these adult ticks (experiment II) were allowed to feed on eight susceptible guinea pigs, which all but one became infected by Ri. rickettsii, as demonstrated by high fever starting at 6–9 DPI, PCR-positive spleen, and/or isolation of rickettsiae from blood by the shell vial technique (Table 2). The rickettsial isolates were confirmed to be Ri. rickettsii by PCR plus DNA sequencing. All guinea pigs that did not present fever were shown to be seronegative for Ri. rickettsii at 21 DPI. Febrile guinea pigs, which were confirmed to be infected by Ri. rickettsii through PCR or shell vial technique, were not tested by immunofluorescence assay. In summary, results of experiments I and II showed that there was transstadial transmission of Ri. rickettsii from larvae to nymphs, and from nymphs to adult ticks, after an initial exposure of uninfected larvae and nymphs to Ri. rickettsii-infected dogs. In addition, both nymphs and adults showed to be competent vectors of Ri. rickettsii to guinea pigs.

Table 2.

Evaluation of Rickettsial Infection in Rhipicephalus sanguineus Ticks and Rickettsial Transmission to Guinea Pigs in Three Experiments

		Infested guinea pigs
						Diagnostic
				Fever ^a		Febrile ^b	stageAfebrile
Exp. (tick stage)	Source (% infection)	No.	No. of ticks/animal	No. with fever	Fever onset (DPI)	PCR/shell vial	IFA ^c	% Infection in the subsequent tick stage
I (N)	G1 (15.2)	6	20	0	—	—	Neg.^d	A (0 [0/9])
I (N)	G2 (7.1)	6	20	1	8	Pos./Pos.	Neg.^d	A (77.8 [7/9])
I (N)	G3 (0)	4	20	0	—	—	Neg.^d	A (0 [0/9])
II (A)	G1 (37.9)	4	10^e	3	6–9	Pos./Pos.	Neg.^d	L (7.8 [7/90])^f
II (A)	G2 (35.8)	4	10^e	4	6	Pos./Pos.	—	L (8.3 [10/120])^f
III (L)	G1 (0)	6	100	0	—	—	Neg.^d	N (0 [0/20])
III (L)	G2 (0)	6	100	0	—	—	Neg.^d	N (0 [0/20])

In Exp. I, unfed nymphs (N) derived from engorged larvae that had fed on G1, G2, or G3 dogs were reared on guinea pigs until the subsequent adult (A) stage. In Exp. II, unfed adults (A) derived from engorged nymphs that had fed on G1, G2, or G3 dogs were reared on guinea pigs until the subsequent larval (L) stage. In Exp. III, unfed larvae (L) derived from engorged females that had fed on G1 or G2 dogs were reared on guinea pigs until the subsequent nymphal (N) stage.

Rectal temperature ≥40°C.

Febrile guinea pigs were euthanized and their blood or/and spleen was submitted to PCR or/and to isolation of Rickettsia in cell culture by the shell vial technique.

Only guinea pigs that did not present fever were tested by IFA.

Serum nonreactive to Ri. rickettsii at the 1/64 dilution.

Five males, five females.

PCR was performed on larvae derived from females that showed to be infected by PCR at the end of oviposition. Larvae were tested in pools, each pool containing five larvae. Results given as Minimal Infection Rate (number of infected pools/total number of larvae tested), which considers at least one infected larvae in each pool.

Exp., experiment; DPI, days postinfestation; PCR, polymerase chain reaction; IFA, immunofluorescence assay; Pos., positive; Neg., negative.

Rickettsia-infected engorged females from experiment II, which had been initially exposed to Ri. rickettsii while feeding as nymphs upon either G1 or G2 dogs, yielded viable eggs that resulted into infected larvae. The overall minimum infection rate of larval offspring was estimated to be 7.8–8.3% (Table 2). Considering individual infected females, 6 larval pools of 5 larvae each were tested per female, totalizing 42 pools (210 larvae) from 7 females. These 7 females were shown to be infected by rickettsiae at the end of oviposition. Only 1–4 larval pools (mean: 2.4) per female were shown to be infected by rickettsia, demonstrating that filial infection rate was generally <50% in the offspring of these females. On the other hand, Table 1 shows that engorged females exposed to Ri. rickettsii-infected dogs while feeding as adults upon either G1 or G2 dogs yielded viable eggs that resulted in no infected larvae, although 100% of the females were shown to be infected at the end of oviposition. As a whole, these results show that rickettsial transovarial transmission occurred when ticks became infected during the nymphal stage, but not when ticks became infected during the adult stage. No tests for transovarial transmission were performed with ticks initially exposed to Ri. rickettsii during the larval stage.

Comparative performance of infected and uninfected tick groups

Tick biological parameters recorded during experiments I–III are shown in Tables 3 –5. Generally, both feeding success (reported as the number of ticks that engorged) and molting success tended to be significantly higher (p < 0.05) for G3 immature ticks (uninfected), or similar between G3 and infected tick groups (G1–G2) in experiments I–III. Additionally, in experiment III both feeding and oviposition success, as well as engorged female weights, were similar (p > 0.05) between G3 and G1–G2 ticks, even though 100% of the G1–G2 engorged females were shown to be infected at the end of oviposition, whereas no G3 female was infected (Table 5). Similarly, oviposition success was also 100% for both G1 and G2 ticks in experiment II (Table 4), when females transmitted rickettsia to part of their larval offspring, after being exposed to feed on guinea pigs.

Table 3.

Biological Parameters of Three Experimental Groups (G1, G2, G3) of Rhipicephalus sanguineus Ticks (Experiment I)

	Experimental groups
Biological parameter	G1 ticks	G2 ticks	G3 ticks
No. of exposed larvae^*	600 [0]	600 [0]	400 [0]
No. of larvae that engorged (%)	528 (88.0)^a	272 (45.3)^b	371 (92.8)^c
Larval feeding period in days^†	4.6 ± 0.7 (4–6)^a	4.5 ± 0.6 (4–6)^a	5.0 ± 0.3 (4–6)^b
Larval molting success (%)	369 (69.9)^a,b	185 (68.0)^a	281 (75.7)^b
No. of exposed nymphs	120 [15.2]	120 [7.1]	80 [0]
No. of nymphs that engorged (%)	64 (53.3)^a	55 (45.8)^a	39 (48.8)^a
Nymphal feeding period in days^†	4.9 ± 0.79 (4–6)^a	5.3 ± 0.71 (4–6)^b	5.4 ± 0.97 (4–7)^b
Nymphal molting success (%)	36 (56.3%)^a	43 (78.2%)^b	22 (56.4%)^a
Unfed adults^*	36 [0]	43 [77.8]	22 [0]

During the larval stage, G1 and G2 groups were exposed to feed on Ri. rickettsii-infected dogs, whereas G3 were exposed to uninfected dogs (control group). Subsequently, nymphs of the three groups were allowed to feed on uninfected guinea pigs, and off-host developmental stages were observed in an incubator at 25°C and 85% relative humidity. Values followed by different superscript alphabetic letters in the same line are significantly different (p < 0.05).

Number within square brackets refer to rickettsial infection rate, as determined by PCR performed on a sample of ticks; see Tables 1 and 2.

†

Values presented as mean ± standard deviation (range in parentheses).

Table 4.

Biological Parameters of Three Experimental Groups (G1, G2, G3) of Rhipicephalus sanguineus Ticks (Experiment II)

	Experimental groups
Biological parameter	G1 ticks	G2 ticks	G3 ticks
No. of exposed nymphs^*	1800 [0]	1800 [0]	1200 [0]
No. of nymphs that engorged (%)	863 (47.9)^a	436 (24.2)^b	1079 (89.9)^c
Nymphal feeding period in days^†	4.8 ± 0.7 (4–8)^a	6.6 ± 0.8 (5–9)^b	5.3 ± 0.5 (4–8)^c
Nymphal molting success (%)	696 (80.1)^a	374 (85.5)^b	892 (82.7)^a,b
No. of exposed females	20 [37.9]	20 [35.8]	Not done
No. of females that engorged (%)	12 (60.0)^a	10 (50.0)^a	—
Feeding period in days^†	9.3 ± 1.4 (7–12)^a	8.6 ± 0.7 (7–9)^b	—
Engorged female weight in mg^†	74.8 ± 45.4 (18.4–150.1)^a	109.8 ± 28.5 (62.5–153.4)^b	—
No. of females that oviposited (%)	12 (100)^a	10 (100)^a	—
Unfed larvae (offspring) tested^*	90 [7.8]	120 [8.3]

During the nymphal stage, G1 and G2 groups were exposed to feed on Ri. rickettsii-infected dogs, whereas G3 were exposed to uninfected dogs (control group). Subsequently, adults of the three groups were allowed to feed on uninfected guinea pigs, and off-host developmental stages were observed in an incubator at 25°C and 85% RH. Values followed by different superscript alphabetic letters in the same line are significantly different (p < 0.05).

Number within square brackets refer to rickettsial infection rate, as determined by PCR performed on a sample of ticks; see Tables 1 and 2.

†

Values presented as mean ± standard deviation (range in parentheses).

Table 5.

Biological Parameters of Three Experimental Groups (G1, G2, G3) of Rhipicephalus sanguineus ticks (Experiment III)

	Experimental groups
Biological parameter	G1 ticks	G2 ticks	G3 ticks
No. of exposed females^*	60 [0]	60 [0]	40 [0]
No. of females that engorged (%)	39 (65.0)^a	55 (91.7)^b	31 (77.5)^a,b
Feeding period (days)^†	7.6 ± 1.1 (7–11)^a,b	9.4 ± 1.2 (8–11)^a	7.7 ± 0.7 (7–9)^b
Engorged female weight (mg)^†	126.8 ± 30.5 (71–201.7)^a	120.7 ± 28.4 (56.4–192.9)^a	117.6 ± 34.5 (51.8–197.2)^a
No. of females that oviposited (%)^*	31 (79.5)^a [100]	55 (100)^b [100]	29 (93.6)^a,b [0]
Egg mass weight (mg) per female^†	75.2 ± 25.4 (10.1–123.5)^a	73.8 ± 19.7 (33–124.6)^a	68.6 ± 26.8 (10.8–117.1)^a
Egg mass% of hatching	>90%	>90%	>90%
CEI^†	56.6 ± 13.0 (9.6–64.6)^a	61.2 ± 4.4 (49.8–77.9)^a	56.3 ± 12.6 (18.8–64.5)^a
No. of exposed larvae^*	600 [0]	600 [0]	Not done
No. of larvae that engorged (%)	321 (53.5)^a	246 (41)^b	—
Larval feeding period (days)^†	4.3 ± 0.53 (3–5)^a	4.5 ± 0.54 (3–5)^a	—
Larval molting success (%)	>90%	>90%	—

During the adult stage, G1 and G2 groups were exposed to feed on Ri. rickettsii-infected dogs, whereas G3 were exposed to uninfected dogs (control group). Subsequently, reproductive parameters of the three groups were observed in an incubator at 25°C and 85% RH, and the larval offspring were allowed to feed on uninfected guinea pigs. Values followed by different superscript alphabetic letters in the same line are significantly different (p < 0.05).

Number within square brackets refer to rickettsial infection rate, as determined by PCR performed on a sample of ticks; see Table 1.

†

Values presented as mean ± standard deviation (range in parentheses).

CEI = mg egg mass/mg engorged female × 100.

Discussion

Earlier studies in Brazil and the United States suggested that domestic dogs could be an important source of Ri. rickettsii infection to ticks (Regendanz and Muniz 1936, Dias 1937, Topping 1947, Price 1954, Keenan et al. 1977). This suggestion was based on epidemiological associations, isolation of Ri. rickettsii from ticks feeding on dogs in endemic areas, and demonstration of the susceptibility of dogs to virulent strains of Ri. rickettsii, evidenced by clinical illness and detectable rickettsemia. Thereafter, Norment and Burgdorfer (1984) exposed dogs to virulent North American strains (Sawtooth and Wachsmuth) of Ri. rickettsii and evaluated if these dogs could serve as source of rickettsial infection for Rh. sanguineus ticks. These authors found that none of the 394 ticks that fed on rickettsemic dogs (infected by needle inoculation) became infected, and only 3 of 348 ticks (0.9% infection rate) were infected after feeding on dogs that had been infected via tick bite (Ri. rickettsii-infected D. andersoni). Based on these data, Norment and Burgdorfer (1984) concluded that the domestic dog, under natural conditions, is not an efficient amplifier host of Ri. rickettsii for ticks.

Differently from the study of Norment and Burgdorfer (1984), our study showed that Ri. rickettsii-needle inoculated dogs served as infection source for 15.2%, 37.9%, and 100% of Rh. sanguineus larvae, nymphs, and adults, respectively, and that dogs infected via tick bite served as infection source for 7.1%, 35.8%, and 100% of the larvae, nymphs, and adults, respectively. As the length of the rickettsemic period reported by Norment and Burgdorfer (1984) in dogs (3–7 days; mean: 5.5 days) was similar to the ones reported for the dogs of our study (3–13 days; mean: 6.3 days) (Piranda et al. 2008), the differences of the infection rates of ticks that fed during the rickettsemic period could be related to rickettsial strain virulence. In this case, our rickettsial strain showed to be more virulent to dogs than the strains used by Norment and Burgdorfer (1984). The six infected dogs of our study became clinically ill with high fever lasting for 6–11 days and developed very high anti-Ri. rickettsii antibody titers (32,768–65,536) (Piranda et al. 2008). In contrast, only two out of the six Ri. rickettsii-infected dogs of the study of Norment and Burgdorfer (1984) became ill with fever, and they all developed low anti-Ri. rickettsii antibody titers (32–256). These differences could be related to higher blood rickettsial concentration in our study, which resulted in higher tick infection rates. In fact, an experimental infection study with several small mammals and D. andersoni ticks showed that high infection rates were obtained only if ticks fed during periods of high rickettsial concentrations in the blood (Burgdorfer et al. 1966). Following these statements, it is also reasonable to speculate that higher infection rates of G1 ticks, when compared with G2 ticks, was a result of higher rickettsial concentrations in the blood of G1 dogs, inoculated intraperitoneally with high infective doses.

We successfully demonstrated transstadial transmission of Ri. rickettsii in Rh. sanguineus ticks (Table 1). In addition, the vector competence of both nymphs (G2 ticks) and adults (G1 and G2 ticks) was demonstrated for susceptible guinea pigs (Table 2). These results are similar to previous studies using guinea pigs in Brazil (Regendanz and Muniz 1936) and using rabbits and guinea pigs in the United States (Parker et al. 1933). Regarding transovarial transmission of Ri. rickettsii in Rh. sanguineus, our results showed that it occurred in engorged females that had been infected as nymphs by feeding on dogs, but not in engorged females that acquired the infection during adult feeding on dogs. Similar results were reported by Parker et al. (1933), who demonstrated transovarial transmission of Ri. rickettsii in Rh. sanguineus engorged females that had been infected as larvae by feeding on infected rabbits or guinea pigs, but no transovarial transmission occurred when the engorged females were first exposed to Ri. rickettsii during adult feeding on infected guinea pigs. According to Burgdorfer and Brinton (1975), efficiency of transovarial transmission of Ri. rickettsii depends primarily on the degree of rickettsial development in ovarian tissues, that is, females with generalized massive infection transmitted rickettsiae to 100% of their progeny; those with lower rickettsial infections or those in which rickettsial development was still in an initial phase at the beginning of oviposition produced considerably lower percentages of infected filial ticks. Applying these statements to our results, it is possible that when Rh. sanguineus females acquired the infection during adult feeding, there was no sufficient time for rickettsiae to colonize the ovaries, precluding transovarial transmission. On the other hand, ovary colonization occurred on time for transovarial transmission when females had been infected as larvae (Parker et al. 1933) or nymphs (this study). Even in the case of nymphs in this study, ovary colonization might have occurred only partially in the adult females, resulting in filial infection rates generally <50%. Further research is needed to clarify this issue.

Our results indicated that Ri. rickettsii strain Taiaçu does not elicit lethal effect on Rh. sanguineus ticks, in contrast to the lethal effect demonstrated by this same strain on A. aureolatum engorged females and their eggs (Labruna et al. 2009). The fact that transovarial transmission of strain Taiaçu is usually accompanied by 100% filial infection rate in A. aureolatum (M.B. Labruna, unpublished data) and <50% filial infection rate in Rh. sanguineus (this study) could be related to more intense multiplication of Ri. rickettsii in A. aureolatum female internal organs, as it has been shown that females with generalized massive infection transmitted rickettsiae to 100% of their progeny (Burgdorfer and Brinton 1975, Burgdorfer 1988). On the other hand, the possible less massive infection in Rh. sanguineus could be the reason for the absence of significant mortality between infected (G1 and G2) and uninfected (G3) engorged female ticks of this study. Finally, it must be stated that Rh. sanguineus is highly adapted to survival on dogs for each life stage, and the guinea pig used in this study might have altered the rate of infection among subsequent tick stages.

In experiment I (Table 2; G2 ticks), infection rates varied from lowest values in unfed nymphs (7.1%) to higher values in unfed adults (77.8%), and then 100% infection rate in engorged females. Similarly, in experiment II (Table 2; G2 ticks), tick infection rates varied from 35.8% in unfed adults to 100% in engorged females. These findings are consistent with transmission between cofeeding ticks on susceptible animals (Philip 1959, Niebylski et al. 1999, Freitas et al. 2009). In this case, a generation of ticks starting with low infection rates among immature stages would result in higher infection rates among adult ticks because uninfected ticks would become infected after cofeeding with infected ticks on susceptible (rickettsemic) hosts during that generation. In nature, this condition certainly does not occur frequently because not all hosts parasitized by ticks are highly susceptible to Ri. rickettsii, as were the guinea pigs used in this study.

During the rickettsemic period, dogs from both G1 (infected via intraperitoneal) and G2 (infected via tick bites) groups were able to infect 15.2–37.9% and 7.1–35.8%, respectively, of the Rh. sanguineus larvae–nymphs that fed on them. As experimental conditions of G2 group is similar to natural conditions (i.e., in which dogs acquire rickettsial infection via tick feeding), our results indicate that dogs can act as amplifier hosts in the field for <45% of the ticks that feed on them during the rickettsemic period. Interestingly, a recent study with the same Rh. sanguineus population (colony Seropédica) used in this study showed that febrile guinea pigs, which had been inoculated intraperitoneally with Ri. rickettsii strain Taiaçu, were able to infect 80–100% of the Rh. sanguineus nymphs that had fed on them during the larval stage (Labruna et al. 2008). These high infection rates indicate that the Rh. sanguineus tick colony used to infest dogs in this study is highly susceptible to Ri. rickettsii. Thus, we can infer that the dogs of our study were ∼10 times less efficient than guinea pigs to infect Rh. sanguineus larvae. Based on these data, our results indicate that rickettsemic dogs can act as amplifier host for Rh. sanguineus, although only a minority of the larvae and nymphs (<45%) feeding on them should become infected by Ri. rickettsii. In addition, it appears that Rh. sanguineus, in the absence of horizontal transmission, would not maintain Ri. rickettsii through successive generations, possibly because of low filial infection rates. The epidemiological significance of our findings needs to be investigated under natural conditions in RMSF endemic areas of Brazil, where besides being infested by Rh. sanguineus, the canine population is continually exposed to A. cajennense and/or A. aureolatum ticks, the main vectors of RMSF to humans in Brazil.

Footnotes

Acknowledgments

The authors thank Laboratório Biovet, Brazil, for providing tick-naïve animals for this study. This work was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo—FAPESP (grant 06/50918-0 to M.B.L.), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior—CAPES (Ph.D. scholarship to E.M.P.), and Conselho Nacional de Desenvolvimento Científico e Tecnológico—CNPq (academic career scholarship to J.L.H.F. and M.B.L.).

Disclosure Statement

Authors declare no conflict of interest.

References

Altschul

, Gish

, Miller

, Myers

et al. Basic local alignment search tool. J Mol Biol, 1990; 215:403–410.

Barrera

. Algunas consideraciones biologicas sobre Rhipicephalus sanguineus (Latreille), 1806, transmisor de fiebre manchada en el norte y noroeste de Mexico. Rev Inst Salubr Enferm Trop, 1956; 16:3–14.

Burgdorfer

. Ecological and epidemiological considerations of Rock Mountain spotted fever and scrub typhus. Walker

. Biology of Rickettsial Diseases, 1. Boca Raton: CRC Inc., 1988; 33–50.

Burgdorfer

, Brinton

. Mechanisms of transovarial infection of spotted fever rickettsiae in ticks. Ann N Y Acad Sci, 1975; 266:61–72.

Burgdorfer

, Friedhoff

, Lancaster

. Natural history of tick-borne spotted fever in the USA. Susceptibility of small mammals to virulent Rickettsia rickettsii. Bull World Health Organ, 1966; 35:149–153.

Bustamante

, Varela

. Distribuicion de las rickettsiasis en Mexico. Rev Inst Salubr Enferm Trop, 1947; 8:3–14.

Cunha

, Fonseca

, Rezende

, Rozental

et al. First identification of natural infection of Rickettsia rickettsii in the Rhipicephalus sanguineus tick, in the state of Rio de Janeiro. Pesqui Vet Bras, 2009; 29:105–108.

Demma

, Traeger

, Nicholson

, Paddock

et al. Rocky Mountain spotted fever from an unexpected tick vector in Arizona. N Engl J Med, 2005; 353:587–594.

Dias

. O cão como provável reservatório do vírus da febre maculosa brasileira. Brasil Medico, 1937; 51:1245–1247.

10.

Dumler

, Walker

. Rocky Mountain spotted fever-changing ecology and persisting virulence. N Engl J Med, 2005; 353:551–553.

11.

Freitas

, Faccini

JLH

, Labruna

. Experimental infection of the rabbit tick, Haemaphysalis leporispalustris, with the bacterium Rickettsia rickettsii, and comparative biology of infected and uninfected tick lineages. Exp Appl Acarol, 2009; 47:321–345.

12.

Gehrke

, Gazeta

, Souza

, Ribeiro

et al. Rickettsia rickettsii, Rickettsia felis and Rickettsia sp. TwKM03 infecting Rhipicephalus sanguineus and Ctenocephalides felis collected from dogs in a Brazilian spotted fever focus in the State of Rio De Janeiro/Brazil. Clin Microbiol Infect, 2009; 15:267–268.

13.

Guglielmone

, Beati

, Barros-Battesti

, Labruna

et al. Ticks (Ixodidae) on humans in South America. Exp Appl Acarol, 2006; 40:83–100.

14.

Horta

, Moraes-Filho

, Casagrande

, Saito

et al. Experimental infection of opossums Didelphis aurita by Rickettsia rickettsii and evaluation of the transmission of the infection to ticks Amblyomma cajennense. Vector Borne Zoonot Dis, 2009; 9:109–117.

15.

Keenan

, Buhles

, Huxsoll

Jr. , Williams

et al. Pathogenesis of infection with Rickettsia rickettsii in the dog: a disease model for Rocky Mountain spotted fever. J Infect Dis, 1977; 135:911–917.

16.

Labruna

. Ecology of rickettsia in South America. Ann N Y Acad Sci, 2009; 1166:156–166.

17.

Labruna

, Ogrzewalska

, Martins

, Pinter

et al. Comparative susceptibility of larval stages of Amblyomma aureolatum, Amblyomma cajennense, and Rhipicephalus sanguineus to infection by Rickettsia rickettsii. J Med Entomol, 2008; 45:1156–1159.

18.

Labruna

, Ogrzewalska

, Moraes-Filho

, Martins

et al. Evaluation of the lethal effect of Rickettsia rickettsii on Amblyomma aureolatum ticks. Presented at the 23rd Meeting of the American Society for Rickettsiology, August 15–18, 2009, Hilton Head Island, South CarolinaAbstract book2009; 29.

19.

Labruna

, Whitworth

, Horta

, Bouyer

et al. Rickettsia species infecting Amblyomma cooperi ticks from an area in the State of São Paulo, Brazil, where Brazilian spotted fever is endemic. J Clin Microbiol, 2004; 42:90–98.

20.

Moraes-Filho

, Pinter

, Pacheco

, Gutmann

et al. New epidemiological data on Brazilian spotted fever in an endemic area of the state of São Paulo, Brazil. Vector Borne Zoonot Dis, 2009; 9:73–78.

21.

Niebylski

, Peacock

, Schwan

. Lethal effect of Rickettsia rickettsii on its tick vector (Dermacentor andersoni) Appl Environ Microbiol, 1999; 65:773–778.

22.

Norment

, Burgdorfer

. Susceptibility and reservoir potential of the dog to spotted fever-group rickettsiae. Am J Vet Res, 1984; 45:1706–1710.

23.

Parker

, Philip

, Jellinson

. Rocky Mountain spotted fever. Potentialities of tick transmission in relation to geographic occurrence in the United States. Am J Trop Med, 1933; 13:241–378.

24.

Philip

. Some epidemiological considerations in Rocky Mountain spotted fever. Public Health Rep, 1959; 74:595–600.

25.

Piranda

, Faccini

, Pinter

, Saito

et al. Experimental infection of dogs with a Brazilian strain of Rickettsia rickettsii: clinical and laboratory findings. Mem Inst Oswaldo Cruz, 2008; 103:696–701.

26.

Price

. The epidemiology of Rocky Mountain spotted fever. II. Studies on the biological mechanism of Rickettsia rickettsii. Am J Hyg, 1954; 60:292–319.

27.

Regendanz

, Muniz

. Pesquisas sobre a transmissão do typho exantematico de S. Paulo por ixodideos. Brasil Medico, 1936; 50:45–48.

28.

Regnery

, Spruill

, Plikaytis

. Genotypic identification of rickettsiae and estimation of intraspecies sequence divergence for portions of two rickettsial genes. J Bacteriol, 1991; 173:1576–1589.

29.

Souza

, Moraes-Filho

, Ogrzewalska

, Uchoa

et al. Experimental infection of capybaras Hydrochoerus hydrochaeris by Rickettsia rickettsii and evaluation of the transmission of the infection to ticks Amblyomma cajennense. Vet Parasitol, 2009; 161:116–121.

30.

Topping

. The epidemiology of Rocky Mountain spotted fever. N Y State J Med, 1947; 47:1585–1587.

31.

Walker

, Bouattour

, Camicas

, Estrada-Pena

et al. Ticks of domestic animals in Africa: a guide to identification of species. Special Publication of International Consortium on Ticks Tick-Borne Disease (ICTTD-2): Atalanta, Houten, The Netherlands, 2003.