Sp

BRPa

b

Dc

d

e

f

g

ARRA

KDEASDPSFA(

1

om

0h

Comparative Immunology, Microbiology and Infectious Diseases 36 (2013) 613– 617

Contents lists available at ScienceDirect

Comparative Immunology, Microbiologyand Infectious Diseases

j o ur na l ho me pag e: www.elsev ier .com/ locate /c imid

tudy of ehrlichiosis in kennel dogs under treatment andrevention during seven months in Dakar (Senegal)

ernard Davousta,∗, Oleg Mediannikovb, Jeanne Chenec, Raphaël Massotc,aphaël Tined, Mamadou Diarrad, Jean-Paul Demoncheauxe,f,ierre Scandolab, Frédéric Beugnetg, Luc Chabannec

Research Unit of Emerging Infectious and Tropical Diseases (URMITE) UMR-CNRS 7278-IRD 198, Marseille, FranceResearch Unit of Emerging Infectious and Tropical Diseases (URMITE) UMR-CNRS 7278-IRD 198, Institute of Research for theevelopment, Dakar, SenegalVetAgro Sup, Veterinary Campus, Marcy l’Etoile, FranceVeterinary Services of the National Gendarmerie, Dakar, SenegalFrench Armed Forces Medical Center in Senegal, Dakar, SenegalAnimal Epidemiology Working Group of the Military Health Service, Toulon, FranceMerial, Lyon, France

a r t i c l e i n f o

rticle history:eceived 13 May 2013eceived in revised form 6 August 2013ccepted 13 August 2013

eywords:oghrlichia canisnaplasma platysenegal

a b s t r a c t

In Dakar kennels where morbidity and mortality attributed to diseases transmitted by tickswere high, we conducted a field study to assess the prevalence of Ehrlichia canis, Anaplasmaplatys and Babesia spp. infections in two kennels (n = 34 dogs) and to study the impact oftick protection. The first day of the study, the E. canis PCR were positive in 18 dogs (53%). A.platys was found in one dog and all dogs were negative for Babesia spp. After one month ofdoxycycline treatment, the number of PCR positive dogs decreased significantly to 2 (5.9%).During seven months, all dogs were treated monthly topically with a novel combination(Certifect®, Merial) delivering at least 6.7 mg fipronil/kg body weight, 8.0 mg amitraz/kgand 6 mg (S)-methoprene/kg. The number of PCR positive dogs remained stable all over

oxycyclineCRerologyipronilmitraz

S)-methoprene

the seven months, with 4 dogs being positive at Day 90 and 2 at Day 210. The combinationof treatment and monthly prevention had a significant effect in the two kennels. All dogsremained healthy, which was not the case in previous years.

© 2013 The Authors. Published by Elsevier Ltd. Open access under CC BY-NC-SA license.

. Introduction

Dakar, capital of Senegal, located on the western edgef Africa, is a city of over one million inhabitants, whereuch of the country’s economic activity is concentrated.

∗ Corresponding author. Tel.: +33 4 91 32 43 75.E-mail address: bernard.davoust@gmail.com (B. Davoust).

147-9571 © 2013 The Authors. Published by Elsevier Ltd. ttp://dx.doi.org/10.1016/j.cimid.2013.08.002

Open access under CC BY

Population growth is constant and requires organizing theprotection of individuals and their households. For this pur-pose, the use of working dogs to prevent delinquency iscommon. The Senegalese Gendarmerie has a special unitin Dakar with handlers and trained dogs. In addition, sev-eral private companies have set up protection units withdogs. These dogs live in kennels in an environment suscep-tible to external parasites especially ticks. As the climate is

tropical, Dakar is exposed to the development of the browndog tick, Rhipicephalus sanguineus. The hot and humid sea-son lasts from June to October with temperatures around27 ◦C and a peak rainfall in August (250 mm) when the

-NC-SA license.

icrobiolo

614 B. Davoust et al. / Comparative Immunology, M

vegetation is abundant. During the dry season, the risk oftick infestation decreases but does not disappear.

R. sanguineus, the brown dog tick, is the mostwidespread species in Dakar, especially in kennels whereit readily colonizes the infrastructure (walls, floors, etc.).It is responsible for the transmission of Ehrlichia canis indogs, the agent of canine monocytic ehrlichiosis. Casesof ehrlichiosis have been already diagnosed by indirectimmunofluorescence (IFA), in Dakar kennels where mor-bidity and mortality were high [1,2].

In this context, it seemed interesting to carry out forthe first time in Senegal, a field study of the prevalenceof E. canis infections and other pathogens in natural con-ditions in the kennels located in Dakar and to study theimpact of tick protection. All guard dogs living in these ken-nels were included. The two kennels and the activity of thedogs were supposed to be different in terms of tick chal-lenge. The dogs were first treated by the administration ofdoxycycline (Ronaxan®, Merial, France) for one month and,simultaneously, the prevention of tick infestation throughthe application of a spot-on (Certifect®, Merial, France) wasput in place for a period of 7 months. The health status butalso the serological and blood PCR results were followedduring a period of 7 months.

2. Materials and methods

2.1. Dogs

In 2012, two groups of working dogs were studied (34dogs in total) living in two kennels located in differentneighbourhoods of Dakar (Fig. 1). The kennels chosen werethe largest in Dakar. Kennel 1, located at Marinas Hann andin which 20 dogs were born in Dakar, belongs to a pri-vate enterprise and is composed of 6 males and 14 females,aged 6 months to 9 years (mean 2.5 years). They were Ger-man Shepherds (5), Belgian Shepherds (2), cross-BelgianShepherds–German Shepherds (9) and cross-Beauce Shep-herds (4). In this group, the dogs were irregularly treatedagainst ticks before the start of the study. Tick infestationwas often reported and clinical cases attributable to ehrli-chiosis were frequently observed (two fatal cases duringthe year preceding the study) (Fig. 2). Dogs live an aver-age of two per enclosure. They are properly fed and caredfor. These dogs always work in industrial areas of Dakar,where they have access to bush areas corresponding to thebiotope of R. sanguineus. They are also located near the seaand have regular access to water. The second group of dogsbelongs to the Senegalese Gendarmerie whose kennel 2 islocated in the district of Mbao. It includes 14 male dogs (12Belgian Shepherds and 2 German Shepherds) aged 2.5–11years (mean 7 years). All dogs were purchased in Belgium(12 dogs), France (1 dog), or Italy (1 dog) and they arrivedin Dakar at the age of two to three years. In this kennel,the dogs were closely monitored. They were groomed dailyand were regularly bathed with acaricides (amitraz) beforethe start of the study. However, cases of tick infestation

were common and infection with E. canis responsible forthe death of a dog shortly before the application of ourprotocol was reported. These dogs hardly go outside theirkennel and have no access to the sea.

gy and Infectious Diseases 36 (2013) 613– 617

2.2. Protocol of the study

The first day of the study (Day 0) and two months later(D 60), all dogs were examined by a veterinarian. Theywere in apparent good health. No tick was found. Bloodsamples (serum and EDTA) were collected from all dogs.The samples with EDTA were transported to the laboratoryin France (VetAgro Sup, Veterinary Campus, Marcy l’Etoile,France) in a cold package. From Day 0 to Day 30, all dogswere treated daily, per os, with doxycycline hyclate (100 mgRonaxan®, Merial, Lyon, France). At Days 0, 30, 60, 90, 120,180 and Day 210 all dogs were treated topically with anovel combination (Certifect®, Merial) based on 2 formu-lations delivering at least 6.7 mg fipronil/kg body weight,8.0 mg/kg amitraz and 6 mg/kg (S)-methoprene. The spot-on formulations were concurrently applied directly to theskin at two locations along the dorsal midline of the neck,one spot in front of the shoulder blades and one at mid-neck. In both kennels, the dogs were regularly followed upfor health purposes. Their handlers examined them dailyfor the possible detection of ticks. In case of presence ofticks, they were preserved in alcohol at 70◦. Blood sampleswere taken from the dogs at days 0, 60, 90, and 180 andcollected in EDTA tubes.

The main goal of the study was to evaluate the feasibilityof the preventive protocol in the natural condition in thetropical Africa. We did not include the placebo-controlledgroup of dogs in the study because of the ethical reasons(working dogs could not be stayed without treatment inthe very aggressive environment).

2.3. Extraction of DNA from samples

DNA was extracted from 200 �L of blood using acommercial DNA extraction kit (QIAamp DNA Mini Kit®)according to the manufacturer’s instructions (Qiagen,Hilden, Germany). DNA was eluted with 80 �L of specificbuffer.

2.4. PCR amplification

A multiplex PCR amplification was performed fromeach DNA blood sample to confirm the presence of dogDNA by using primers that specifically amplify a 290-bpfragment of the mitosin gene of Canis familiaris, as previ-ously described [3]. This amplification is used to valid DNAextraction. The presence of one or more pathogens was ableto be detected by using primers: a 380-bp fragment of virB9protein gene of E. canis, a 619-bp fragment of 18S rRNAgene of Babesia or Theileria spp as previously described [4],and a 515-bp fragment of groEL gene of Anaplasma platysas previously described [5]. The amplification reaction wascarried out in a thermocycler (Biometra® T gradient, Goet-tingen, Germany) in a total of 25 �L containing 1.25 �L of2 �M of each primer, 12.5 �L of 2x Type-it Multiplex PCRMaster mix (Qiagen® Multiplex PCR Kit, Qiagen, Hilden,Allemagne), 1 �L DNA and 9 �L of RNase free water. The

following conditions were applied: 95 ◦C 5 min, 35 cycles at95 ◦C 30 s, 61 ◦C 90 s, 72 ◦C 30 s and 60 ◦C 30 min. A negativecontrol (reaction mix without DNA) and a positive control(mix of pathogens DNA and dog DNA) were systematically

B. Davoust et al. / Comparative Immunology, Microbiology and Infectious Diseases 36 (2013) 613– 617 615

study a

iwbu

2

rEFcsTwvtc

3

hTtr

Fig. 1. Localization of the

ncluded in parallel. An aliquot of each amplified productas separated on 1.5% agarose gel stained with Ethidium

romide at 130 V for 1 h. The bands were visualized underltraviolet illumination.

.5. Serology

The sera were tested at Days 0, 60, 90 and 210 with oneapid immuno-migration (RIMTM) quick test that detects. canis antibodies (Witness® Ehrlichia, Synbiotics, Lyon,rance). Sensitized colloidal gold particles bound to anti-E.anis antibodies present within the sample (whole blood,erum or plasma) are allowed to migrate along a strip.he complex is then captured on a sensitized reaction linehere its accumulation causes the formation of a clearly

isible pink band. A control band, located at the end ofhe reading window, ensures that the test was performedorrectly.

. Results

During the 6-month study period, all dogs remained

ealthy, which was not the case in previous years (Fig. 2).he original treatment of the dogs and the monthly anti-ick prevention provided a complete control of any clinicalesurgence of ehrlichiosis.

Fig. 2. Deaths and cases attributed to infec

nd the two dog kennels.

During the survey we were able to collect ticks fromdogs (only in kennel 1). In total, 45 ticks were found from12 dogs. During the first month of treatment we collected31 ticks (including 9 engorged); the other 14 non engorgedticks were found randomly during the next six months. Allticks were morphologically identified as R. sanguineus tickgroup adults using standard taxonomic keys for adult ticksby one of us (FB). In kennel 2, no tick had been observedin the dogs during the 7 months of daily follow-up. Basedon the mode of life of the dogs in both kennels, especiallyknowing that kennel 1 dogs regularly go outside, it washypothesized that the tick challenge was higher in kennel1.

The seroprevalence of ehrlichiosis was 85% at the startof the study (Table 1). After 1-month treatment, the sero-prevalence remained at 85% and was even at 94% at the endof the 7 months. The 30 days of doxycycline had no impacton the serological status in either kennel. There is also nostatistical difference between the two groups and betweenseroprevalences all along the 7 months (p = 0.98).

The first day of the study (Day 0), the PCR were posi-tive in 18 dogs (53%) (Table 1). A. platys was found by PCR

analysis in one dog (also positive for E. canis). All other PCRwere positive for E. canis only. PCR examinations for Babesiaspp. were negative in all cases. After one month of doxycy-cline treatment, the number of PCR positive dogs decreased

tious diseases in kennels 1 and 2.

616 B. Davoust et al. / Comparative Immunology, Microbiology and Infectious Diseases 36 (2013) 613– 617

Table 1Serology and molecular detection of Ehrlichia canis during seven months.

Kennel Localization in Dakar No. of dogs inthe study

Detection of Ehrlichia canisa

D 0 D 60 D 90 D 210

PCR Serologyb PCR Serologyb PCR Serologyb PCR Serologyb

1 Hann Marinas 20 9(45%) 15 (75%) 2 (10%) 15 (75%) 4 (20%) 17 (85%) 2 (10%) 18 (90%)2 Mbao 14 9(64%) 14 (100%) 0 (0%) 14 (100%) 0 (0%) 14 (100%) 0 (0%) 14 (100%)

Total 34 18(53%) 29 (85%) 2 (6%) 29 (85%) 4 (12%) 31 (91%) 2 (6%) 32 (94%)

a Positive samples.b Rapid immuno-migration (Witness® Ehrlichia).

significantly to 2 (5.9%) (p < 2 × 10−5), indicating a reduc-tion of 88.88%. This number of PCR positive dogs remainedstable all over the 7 months, with 4 dogs being positive atDay 90 and 2 at Day 210. The combination of treatmentand monthly prevention had a significant effect in kennel1 (p < 0.02) and kennel 2 (p < 10−7). There was no statisticaldifference in effect between kennel 1 and 2 despite a trendfor a better protection seen in kennel 1.

We have recorded the differences in the dynamics of thedetection of E. canis between two kennels. In the kennel 2all dogs that were positive in the beginning of the studybecame negative from the Day 90. However, in the kennel1, only 6 of 11 dogs became PCR-negative during the studyfrom the Day 90. 2/11 also became negative, but only fromDay 210. Another 2 dogs, vice versa, became positive byPCR beginning from the Day 90 and one initially positivewas found negative at the Day 90 and positive at the Day210.

4. Discussion

Canine monocytic ehrlichiosis (CME) has been reportedin Africa (Congo, Sudan, Niger, South Africa, Tunisia,Kenya). CME is transmitted by the brown dog-tick (R. sang-uineus). It is ubiquitous below the 45th parallel north andcorresponds to the area of distribution of this tick. Thisspecies is especially endophilic, it is often found in ken-nels and crevices of walls. The seroprevalence of CME isvery high throughout Africa where it may be consideredenzootic. The prevalence of E. canis in R. sanguineus ticks inAfrican countries was reported in Cameroon with 6 to 21%of PCR+ ticks [6] and in Ivory Coast with 7 to 47% in tickscollected from watchdogs in Abidjan [7] .The author previ-ously found that over 77% of dogs in eight farms (64 dogs) inAbidjan (Ivory Coast) became seropositive during the firstyear [8]. Native African dogs are often asymptomatic, whileimported dogs that do not have acquired immunity oftendevelop severe clinical forms. The results of this study (85%of seroprevalence) are comparable to those we obtainedwith IFA test in Libreville (Gabon) (83% of 80 dogs) [9]. Inkennel 2 of the Senegalese Gendarmerie, all dogs (100%)were seropositive whereas in the same kennel, in 1990,the seroprevalence rate was already high (78%) [1]. The

persistence of anti-E. canis antibodies after treatment iswell known, but it is not certain whether the antibodiesare markers of chronic infection [10] or of regular reinfec-tions. The fact that ticks were observed may be in favour of

regular antigen contact and reboost for antibody produc-tion. It indicates that the anti-tick prevention was not ableto avoid any antigen contact, therefore the attachment of afew ticks on dogs.

The PCR is currently used to estimate the prevalence ofDNA of E. canis in dog’s blood. In Cameroon, 14.4% (15/104)of dogs were positive and all had attached ticks [11]. InSudan, the prevalence estimated by molecular methodswas 80.8% (63/78) among village dogs, and R. sanguineuswere found on all of them [12].

Our study is one of the rare field studies in the tropi-cal Africa in the natural conditions. The main result of ourstudy is the approval of the feasibility of the applicationof the basic prevention protocol in the natural conditionand the confirmation of already known protective effectsof the Certifect®. Our results indicate also that appropriatedoxycycline treatment eliminates E. canis and A. platys DNAas measured by PCR. It has been shown that PCR is a goodindicator of the persistence of E. canis in different tissues ofthe dog [13]. Experimentally, it has been shown that PCRbecomes negative in the blood [14]. However, we also knowthat the infection may persist even after 6 weeks of treat-ment with doxycycline [15]. The most recent experimentalstudies show that one month with doxycycline treatmentproduces a negative PCR in blood in the long-term (211days), even after immunosuppression [16].

Regular anti-tick treatments are known to help indecreasing the risk of disease transmission. Davoust et al.[17] and Otranto et al. [18] demonstrated respectively thatthe use of Frontline® and Advantix® gave a certain levelof protection to dogs against E. canis infection in field tri-als in Africa and in Italy. It was demonstrated recently thatCertifect® induced an indirect protection against the trans-mission of E. canis using an experimental model based ona strain of E. canis infected R. sanguineus [19]. The com-bination of fipronil and amitraz increases the speed ofkill and disrupts the attachment of ticks, therefore pro-viding a better preventive effect against the transmissionof pathogens by ticks than fipronil alone. In this 7-monthstudy, Certifect® provided a 88.9% significant protection inkennel dogs with a PCR positive rate decreasing from 53%to 6% (p < 2 × 10−5). This protection was 77.77% in kennel1 (p = 0.013) and 100% in kennel 2 with a percentage of

PCR dogs decreasing from 64% to 0% (p = 0.00027), whichdemonstrates that several factors may affect the protectiveefficacy, especially the level of tick challenge. Effectively,the different results observed in the two kennels in our

icrobiolo

sTsttwm

mpii

C

Leip

A

ta

R

[

[

[

[

[

[

[

[

[

B. Davoust et al. / Comparative Immunology, M

tudy are due to the fact that kennel 1 was very infested.he lifestyle of dogs of this kennel is also a factor of expo-ure to ticks because of the random contact with differentick-infested areas and frequent changes of handlers. Fur-hermore, these dogs were probably taken to sea bathshich could reduce the effectiveness of preventive treat-ent.Overall, this study shows that the doxycycline treat-

ent eliminates the ehrlichia carriage in dogs and thatreventive treatment protects dogs from the tick-borne

nfections, reducing the mortality and morbidity due tonfectious diseases.

onflict of interest statement

The work reported herein was funded by Merial,yon, France. One author (Fréderic Beugnet) is currentlymployee of Merial, however there were no conflictingnterests that may have biased the work reported in thisaper.

cknowledgments

We thank Sérigne Abdoulaye Cissé for the assistance onhe description of clinical symptoms, Mamadou Diedhiound Masse Sambou for technical help.

eferences

[1] Parzy D, Davoust B, Raphenon G, Vidor E. L’ehrlichiose canine auSénégal: enquête séroépidémiologique humaine et canine à Dakar.Méd Trop 1991;51(1):59–63.

[2] Brouqui P, Davoust B, Haddad S, Vidor E, Raoult D. Serological evalua-tion of Ehrlichia canis infections in military dogs in Africa and ReunionIsland. Vet Microbiol 1991;26:103–5.

[3] Criado-Fornelio A, Gónzalez-del-Río MA, Buling-Sarana A, Barba-Carretero JC. Molecular characterization of a Babesia gibsoni isolatefrom a Spanish dog. Vet Parasitol 2003;117:123–9.

[4] Kledmanee K, Suwanpakdee S, Krajangwong S, Chatsiriwech J, Suk-sai P, Suwannachat P, et al. Development of multiplex polymerase

chain reaction for detection of Ehrlichia canis, Babesia spp. and Hep-atozoon canis in canine blood. Southeast Asian J Trop Med PublicHealth 2009;40:35–9.

[5] Beall MJ, Chandrashekar R, Eberts MD, Cyr KE, Diniz PP,Mainville C, et al. Serological and molecular prevalence of Borrelia

[

gy and Infectious Diseases 36 (2013) 613– 617 617

burgdorferi, Anaplasma phagocytophilum, and Ehrlichia species in dogsfrom Minnesota. Vector Borne Zoonotic Dis 2008;8:455–64.

[6] Ndip LM, Ndip RN, Ndive VE, Awuh JA, Walker DH, McBride JW.Ehrlichia species in Rhipicephalus sanguineus ticks in Cameroon. Vec-tor Borne Zoonotic Dis 2007;7:221–7.

[7] Socolovschi C, Gomez J, Marié JL, Davoust B, Guigal PM, Raoult D,et al. Ehrlichia canis in Rhipicephalus sanguineus ticks in the IvoryCoast. Ticks Tick Borne Dis 2012;3:410–2.

[8] Davoust B, Gomez J, Colombet G, Marié JL, Guigal PM. Ehrlichiosemonocytaire canine: précocité de l’infection à Abidjan (Côte d’Ivoire).In: Proceedings of the IVe Congrès international d’épidémiologie del’Association des épidémiologistes de langue franc aise (ADELF) et del’Association pour le développement de l’épidémiologie de terrain(EPITER) et XVIes Actualités du Pharo. 2010. p. 5–7.

[9] Davoust B, Vedy S, Bourry O, Daries W, Casali F, Parzy D, et al. Enquêtede prévalence sanguine de l’ehrlichiose, de la borréliose de Lyme,de la babésiose et de la dirofilariose chez le chien à Libreville et àPort-Gentil (Gabon). Proceedings of the XIIIes Actualités du Pharo,Marseille, France. Méd Trop 2007;67:391.

10] Harrus S, Waner T. Diagnosis of canine monocytic ehrlichiosis(Ehrlichia canis): an overview. Vet J 2010;187:292–6.

11] Ndip LM, Ndip RN, Esemu SN, Dickmu VL, Fokam EB, Walker DH,et al. Ehrlichial infection in Cameroonian canines by Ehrlichia canisand Ehrlichia ewingii. Vet Microbiol 2005;111:59–66.

12] Inokuma H, Oyamada M, Davoust B, Boni M, Dereure J, BuchetonB, et al. Epidemiological survey of Ehrlichia canis and relatedspecies infection in dogs in eastern Sudan. Ann NY Acad Sci2006;1078:461–3.

13] Gal A, Loeb E, Yisaschar Y, Baneth G. Detection of Ehrlichia canis byPCR in different tissues obtained during necropsy from dogs sur-veyed for naturally occurring canine monocytic ehrlichiosis. Vet J2008;175:212–7.

14] Harrus S, Kenny M, Miara L, Aizenberg I, Waner T, Shaw S. Com-parison of simultaneous splenic sample PCR with blood sample PCRfor diagnosis and treatment of experimental Ehrlichia canis infection.Antimicrob Agents Chemother 2004;48:4488–90.

15] Harrus S, Waner T, Aizenberg I, Bark H. Therapeutic effect of doxy-cycline in experimental subclinical canine monocytic ehrlichiosis:evaluation of a 6-week course. J Clin Microbiol 1998;36:2140–2.

16] Gaunt SD, Beall MJ, Stillman BA, Lorentzen L, Diniz PPVP, Bre-itschwerdt EB. Experimental infection and co-infection of dogs withAnaplasma platys and Ehrlichia canis: hematologic, serologic andmolecular findings. Parasit Vectors 2010;3:33.

17] Davoust B, Marié JL, Mercier S, Boni M, Vandeweghge A, Parzy D, et al.Assay of fipronil to prevent canine monocytic ehrlichiosis in endemicareas. Vet Parasitolol 2003;112:91–100.

18] Otranto D, Paradies P, Testini G, Latrofa MS, Weigl S, Cantacessi C,et al. Application of 10% imidacloprid/50% permethrin to preventEhrlichia canis exposure in dogs under natural conditions. Vet Par-

asitolol 2008;153:320–8.

19] Fourie J, Ollagnier C, Beugnet F, Luus G, Jongejan F. Preventionof transmission of Ehrlichia canis by Rhipicephalus sanguineus ticksto dogs treated with a combination of fipronil, amitraz and (S)-methoprene (CERTIFECT®). Vet Parasitol 2013;193:223–8.