O

Ad

IRa

b

c

d

a

ARRA

KADSII

I

ad2slis

S

M3

1h

Ticks and Tick-borne Diseases 4 (2013) 83– 88

Contents lists available at SciVerse ScienceDirect

Ticks and Tick-borne Diseases

jo u rn al hom epage: www.elsev ier .com/ locate / t tbd is

riginal article

ssociation between Anaplasma phagocytophilum seroprevalence in dogs andistribution of Ixodes ricinus and Ixodes persulcatus ticks in Latvia�

nese Berzinaa,∗, Valentina Capliginac, Antra Bormaneb, Agne Pavulinac, Viesturs Baumanisc,enate Rankac, Rita Grantad, Ilze Matisea

Latvia University of Agriculture, Faculty of Veterinary Medicine, Jelgava, LatviaInfectiology Center of Latvia, Riga, LatviaLatvian Biomedical Research and Study Center, Riga, LatviaBIOR, Riga, Latvia

r t i c l e i n f o

rticle history:eceived 25 January 2012eceived in revised form 3 August 2012ccepted 8 August 2012

eywords:. phagocytophilumogeroprevalence. ricinus. persulcatus

a b s t r a c t

Anaplasma phagocytophilum has been detected in ticks in Latvia; however, this is the first study to inves-tigate this pathogen in dogs in Latvia. The aims of this study were: (i) to determine A. phagocytophilumseroprevalence in dogs, (ii) to correlate A. phagocytophilum seroprevalence in dogs with the geographicdistribution of the tick species Ixodes ricinus and Ixodes persulcatus, and (iii) to determine if seroprevalencefor A. phagocytophilum is higher in dogs with clinical signs suggestive of canine granulocytic anaplasmo-sis (CGA). Peripheral venous blood samples were collected from 3 dog groups: (i) clinically healthy dogs(HD, n = 400), (ii) clinically healthy hunting dogs (HHD, n = 41), and (iii) dogs with a clinical suspicionof anaplasmosis (SD, n = 29). Sampling was carried out in regions inhabited by I. ricinus (IR), I. persul-catus (IP), and in regions where both tick species were present (M). SNAP 4Dx test (IDEXX) was usedto detect antibodies against A. phagocytophilum in the blood of all dogs; nested PCR was performed inselected dogs of the SD group. Seroprevalence for A. phagocytophilum was calculated and correlated withthe prevalent tick species in the region. A. phagocytophilum seroprevalence was 11.0% in HD, 12% in

HHD, and 17% in SD with no significant differences among groups. In the IR region, seroprevalence was12.5% (34/272) while seroprevalence in the M region was 17% (13/76), and both were significantly higherthan the seroprevalence of 2% in the IP region (2/93; p < 0.0005). One CGA case was diagnosed. We con-clude that A. phagocytophilum seroprevalence in Latvia is within the range reported from other Europeancountries. CGA should be included in the differential list in Latvian dogs with appropriate clinical signsand laboratory abnormalities, especially in I. ricinus habitat areas.

ntroduction

Tick-borne diseases (TBD), especially tick-borne encephalitisnd Lyme borreliosis have been recognized as important infectiousiseases in humans in Latvia (Bormane et al., 2004; Ranka et al.,004; Süss, 2011). More recently, human granulocytic anaplasmo-is cases have been diagnosed in Latvia and reported in the scientific

iterature (Bormane, 2007). Until now, there have been no stud-es performed to evaluate the presence of any TBD in any animalpecies in Latvia. There have been unpublished and unconfirmed

� Abstract was presented at the International Meeting of Emerging Diseases andurveillance, Vienna, Austria, March 2011.∗ Corresponding author at: Latvia University of Agriculture, Faculty of Veterinaryedicine, Preclinical Institute, Pathology Dept., Kr. Helmana Street 8, Jelgava LV-

004, Latvia.E-mail address: inese.berzina@gmail.com (I. Berzina).

877-959X/$ – see front matter © 2012 Elsevier GmbH. All rights reserved.ttp://dx.doi.org/10.1016/j.ttbdis.2012.08.003

© 2012 Elsevier GmbH. All rights reserved.

case reports of canine babesiosis and borreliosis that have beendiscussed in the local veterinary community.

Canine granulocytic anaplasmosis (CGA), caused by an obliga-tory intracellular rickettsia, Anaplasma phagocytophilum, has beendiagnosed worldwide, including Scandinavia and central Euro-pean countries adjacent to Baltic countries (Latvia, Lithuania, andEstonia) (Egenvall et al., 2000a; Melter et al., 2007; Stuen, 2007;Carrade et al., 2009; Kybicova et al., 2009; Kohn et al., 2011). Infec-tion with A. phagocytophilum in dogs is mostly asymptomatic orcharacterized by nonspecific clinical signs such as fever, lethargy,anorexia, and painful joints (Melter et al., 2007; Carrade et al., 2009;Silaghi et al., 2011). The most common laboratory abnormalitiesnoted in sick dogs are thrombocytopenia, anemia, and increasedliver enzymes (Ravnik et al., 2011). Carrade proposed that the dis-

tribution of CGA parallels that of the human disease; thus, wehypothesized that this disease might be present in dogs in Latvia,but not diagnosed because of the lack of awareness in the veterinarycommunity (Carrade et al., 2009).

84 I. Berzina et al. / Ticks and Tick-bor

Fig. 1. I. ricinus-inhabited region (dark gray), I. persulcatus-dominated region(white), and mixed tick species regions (light gray). Numbers represent the sero-prevalence of A. phagocytophilum in each of the tick regions. Tervete, Skrunda, andLb(

nAesow

sstmc

M

D

tpnHi(dt

ctSgcwibvpitbn

a

imbazi were high-seroprevalence locations within the I. ricinus habitat area. Neigh-oring countries of Latvia are Estonia (EE), Russia (RU), Belarus (BY), and LithuaniaLT).

The main A. phagocytophilum vector tick in Europe is Ixodes rici-us while Ixodes persulcatus has been reported as the main vector insia and Russia (Parola and Raoult, 2001; Bormane, 2007; Carradet al., 2009; Paulauskas et al., 2009; Rar et al., 2011). Both tickpecies are present in Latvia, and there are distinct regions wherenly either I. ricinus or I. persulcatus are prevalent as well as regionshere both tick species are present (Bormane et al., 2004).

The aim of this study was (i) to determine A. phagocytophilumeroprevalence in dogs in Latvia, (ii) to correlate A. phagocytophilumeroprevalence in dogs with the geographic distribution of the 2ick species I. ricinus and I. persulcatus in Latvia, and (iii) to deter-

ine if A. phagocytophilum seroprevalence is higher in dogs withlinical signs suggestive of CGA.

aterials and methods

ogs

Blood samples from dogs were obtained from November 2009ill July 2011. Three groups of dogs were established for the pur-ose of the study: (i) clinically healthy pet dogs (healthy dogs, HD;

= 400), (ii) clinically healthy hunting dogs (healthy hunting dogs,HD; n = 41), and (iii) dogs with a suspicion of A. phagocytophilum

nfection based on clinical signs and abnormal laboratory resultssick dogs, SD; n = 29). For some analyses, the 2 groups of healthyogs were combined. In the text, this combined group is referredo as HD + HHD (n = 441).

HD and HHD were recruited for the study at various veterinarylinics with the help of local veterinarians (dogs brought in for rou-ine vaccinations, scheduled surgeries, or solicited for the study).ample collection from HD was planned to represent various geo-raphic regions of Latvia; however, a larger number of samples wasollected from more densely populated areas (Fig. 1). Blood samplesere also collected to represent geographic regions predominantly

nhabited by I. ricinus or I. persulcatus and from regions inhabited byoth tick species (Fig. 1). Data on the abundance of the tick species inarious geographic regions of Latvia was used from the previouslyublished study by Bormane et al. (2004). The country was divided

nto 3 areas of tick habitats: (i) inhabited by I. ricinus (IR), (ii) inhabi-ed by I. persulcatus (IP), both at >90% proportion, and (iii) inhabited

y both tick species (M) with both species at approximately equalumbers.

To select dogs for the SD group, veterinary practitioners weresked to collect blood samples from dogs that had clinical signs

ne Diseases 4 (2013) 83– 88

of CGA and to submit them to the primary researcher (IB). Casedefinition for A. phagocytophilum infection that warranted inclu-sion of the dog into the SD group included combinations of thefollowing clinical signs and laboratory test abnormalities: fever,lethargy, anorexia, painful joints, lameness, anemia, thrombocy-topenia of unknown cause (Jensen et al., 2007; Kohn et al., 2008,2011). Dogs were included in the study regardless of having a his-tory of tick exposure. Dogs, females and males, various age groups,and various breeds were included in the study.

At the time of sample collection, all dog owners filled out a con-sent form and a questionnaire regarding the use of tick repellentsand other anti-ectoparasite treatments. Additionally, informationwas obtained regarding travel history outside of Latvia, and dogswith history of traveling were excluded from this study. The studyplan was approved by the Animal Protection and Ethics Committeeof the Latvian State Food and Veterinary Service.

Serology and hematology

Two millilitres of peripheral blood were collected in EDTA anti-coagulant tubes, and fresh blood smears were made. The bloodsmears were air-dried and stained (Rapid Differential Stain Kit,VetOne, USA). Blood samples were stored in cooled conditionsand examined with a hematology analyzer (BC-2800VET, DiamondDiagnostics, USA) within 3 days of sampling. Antibody testing wasdone on EDTA whole blood using a SNAP 4Dx test (IDEXX Labo-ratories, Westbrook, Maine, USA) according to the manufacturer’sinstructions available at the website idexx.com. This test detectsIgG and IgM antibodies against p44/msp2 of A. phagocytophilum,antibodies for Ehrlichia canis proteins p30 and p30-1, and C6 pep-tide of Borrelia burgdorferi sensu lato and antigen of Dirofilariaimmmitis. The study regarding Borrelia is ongoing and will be pub-lished separately. Microscopical examination was carried out, and200 granulocytes were counted (Bakken et al., 1996) to verify theresults produced by the hematology analyzer and to screen forintracellular morulae or other clinically significant abnormalitiesthat would indicate the need for treatment or further diagnosticwork-up.

DNA isolation for PCR

Blood samples from SD with thrombocytopenia (<50 × 109/L)and/or with morulae in the peripheral blood granulocytes werefrozen at −20 ◦C, subjected to DNA extraction and subsequentlyanalyzed for A. phagocytophilum-specific nucleotide sequence bynested PCR. EDTA whole blood (2 ml) was centrifuged at 4000 rpmfor 15 min at 4 ◦C. The pellet was lysed with 4 times the volumeof lysis buffer [0.32 M sucrose, 10 mM Tris–HCl (pH 7.6), 5 mMMgCl2, 1% Triton-X-100] for 15 min at 4 ◦C, then centrifugation wasrepeated as described above. The pellet was washed with an equalvolume of lysis buffer, and the centrifugation step was repeated.The pellet was dissolved in 5 ml of cell suspension solution [25 mMEDTA (pH 8.0), 75 mM NaCl] to which 0.5 ml 10% sodium dodecylsulfate and 4.5 �l proteinase K (600 U/ml, 20 mg/ml; Fermentas LifeSciences, Lithuania) were added for cell digestion (1 h at 50 ◦C inwater bath). DNA extraction was performed once with an equalvolume of buffered phenol (pH 8.0) and once with chloroform. TheDNA was precipitated after the addition of 0.6 ml of isopropanoland was collected by centrifugation at 4000 rpm for 10 min at 4 ◦C.The pellet was washed with ice-cold 70% ethanol, dried, dissolved in

100 �l 1× TE-buffer (10 mM Tris–Cl, 1 mM EDTA, pH 8.0) and storedat −20 ◦C, until it was used as a template for PCR amplification. TheDNA concentration was measured in a spectrophotometer (ND-1000 UV-VIS Spectrophotometer, NanoDrop Technologies, USA),

I. Berzina et al. / Ticks and Tick-bor

Table 1Oligonucleotide primers used in nested PCR for A. phagocytophilum 16S rRNA genepartial amplification in this study.

Primer Sequence (5′–3′) Reference

ge3a Outer forward CACATGCAAGTCGAACGGATTATTC Liz et al. (2000)ge10r Outer reverse TTCCGTTAAGAAGGATCTAATCTCC Liz et al. (2000)

ao

P

nsiAmaPidpbotaWIwibmC

D

tsl(m3gDADArcM

S

ctiapra

ge9f Inner forward AACGGATTATTCTTTATAGCTTGCT Liz et al. (2000)ge2 Inner reverse GGCAGTATTAAAAGCAGCTCCAGG Liz et al. (2000)

nd quality verification was performed by running 200 ng of DNAn 0.8% agarose gel (Kohn et al., 2011).

CR amplification

For amplification of the 16S rRNA gene of A. phagocytophilum, aested PCR was performed as published by Liz et al. (2000), withome modifications. The sequences of the primers used are listedn Table 1. All primers were synthesized by Metabion InternationalG, Germany, and all reagents for PCR were purchased from Fer-entas Life Sciences, Lithuania. Amplifications were performed in

thermal cycler Mastercycler epgradientS (Eppendorf, Germany).CRs were performed in 50 �l of the reaction mixture contain-ng 1× Taq Buffer with (NH4)2SO4, 1.5 mM MgCl2, 200 �M of eachNTP, 0.5 �M outer primers for primary reactions or 0.2 �M innerrimers for nested reactions, 1.5 U of Taq DNA polymerase (recom-inant), and 2 �l of DNA template for primary reactions or 2 �lf the primary PCR products for nested reactions. Cycling condi-ions were the same as described by Liz et al. (2000), except thatnnealing was performed at 58 ◦C. The DNA of A. phagocytophilumebster strain (kindly provided by Dr. Friederike von Loewenich,

nstitute of Medical Microbiology, University of Freiburg, Germany)as used as a positive control. Amplified products were visual-

zed by electrophoresis in a 2% agarose gel with 0.2 �g of ethidiumromide/ml by transillumination with a UV light. Quality controleasures included those described previously (Liz et al., 2000;

anelas Domingos et al., 2011).

NA sequencing and phylogenetic analysis

The PCR-amplified 16S rRNA gene fragment of A. phagocy-ophilum was sequenced with the forward primer ge9f. Theequencing reaction was carried out in 25 cycles under the fol-owing conditions: 94 ◦C for 30 s, 58 ◦C for 15 s, and 60 ◦C for 4 minMastercycler epgradient S, Eppendorf, Germany). The sequenced

aterial was analyzed by standard technique using an ABI Prism100 Genetic Analyzer (Perkin-Elmer, USA). The DNA sequenceenerated in the present study was aligned and compared to theNA sequences in GenBank database using the BLAST (Basic Locallignment Search Tool) program to verify that A. phagocytophilumNA was amplified (Altschul et al., 1990). A phylogenetic tree of. phagocytophilum was constructed based on the alignment of 16SRNA gene sequences obtained from the CGA dog and from ticksarrying A. phagocytophilum in Latvia by sequence analysis softwareEGA5 (Tamura et al., 2011).

tatistical analysis

Student’s t-test was used to determine differences in A. phago-ytophilum seroprevalence among HD, HHD, and SD groups of dogs.-Test was also used to compare A. phagocytophilum seroprevalencen HD+HHD dogs in the 3 types of regions defined by tick species

bundance (IR, IP, and M regions). For hematology parameters, non-arametrical distribution of data was assumed, and median andange were calculated as follows: dog groups HD + HHD seroneg-tive, HD + HHD seropositive, SD seronegative, SD seropositive.

ne Diseases 4 (2013) 83– 88 85

Differences between the hematology parameters in these groupswere assessed by Mann–Whitney U test. The percentage of theresults outside (below and above) the reference interval was cal-culated. Values of p < 0.05 were considered significant (NCSS 2007,Utah, USA).

Results

Demographic characteristics of dogs

From November 2009 to July 2011, 470 canine blood sampleswere collected from dogs representing HD, HHD, and SD groups.Information about the age and sex of the dogs included in eachgroups is summarized in Table 2. In all dog groups, seropositivedogs tended to be older compared to seronegative dogs (Table 2).

Dogs from various breeds were included in the study. The HDgroup contained mixed-breed dogs (n = 156), German shepherds(n = 53), dachshunds (n = 30), Rottweilers (n = 13), Russian Europeanlaikas (n = 12), Labrador retrievers (n = 11), French bulldogs (n = 9),boxers (n = 9), Bernese mountain dogs (n = 8), and pugs (n = 7).Thirty-nine other breeds were represented by 5 or fewer dogs.

In the HHD group, the following breeds were represented:Russian European laikas (n = 14), dachshunds (n = 8), Bavarianbloodhounds (n = 6). Seven other breeds were represented by fewerthan 5 dogs.

The SD group had mixed-breed dogs (n = 5) and German shep-herd dogs (n = 4). Thirteen other breeds were represented by 1 or 2dogs each. The breed of 5 dogs was not known.

Hematology

Leukocyte, erythrocyte, and platelet counts are listed in Table 3.In the HD + HHD and SD groups, leukocytosis and thrombocyto-penia were the most commonly observed alterations, followed byanemia of variable severity. Upon microscopical examination, noclinically significant abnormalities were noted in blood smearsof the HD + HHD dogs. Microscopical findings in the SD groupincluded severe anemia, agglutination, spherocytes, leukocytosiswith regenerative left shift, and mild to moderate toxic change inneutrophils. No statistically significant differences in cell countswere noted between HD + HHD-seronegative and seropositive andSD-seronegative and seropositive dogs (both p > 0.05).

A. phagocytophilum in healthy dogs

Seroprevalence of A. phagocytophilum in the HD group was 11.0%(44/400), and in the HHD group, it was 12% (5/41). The differ-ence between the groups was not statistically significant (p = 0.833).Demographical information about seropositive dogs from thesegroups is presented in Table 2. Four HD dogs that were seroposi-tive for A. phagocytophilum were also seropositive for B. burgdorferisensu lato.

Seroprevalence for A. phagocytophilum in the HD + HHD groupfrom western and central Latvia inhabited primarily by I. rici-nus ticks (IR) was 12.5% (34/272), while in the eastern region ofLatvia inhabited primarily by I. persulcatus ticks (IP), seropreva-lence was 2% (2/93). In regions inhabited by both tick species (M),the seroprevalence was 17% (13/76). Seroprevalence in dogs wasnot statistically different between IR and M regions (p = 0.300), butstatistically significant differences were detected between IR andIP regions (p = 0.003) and between M and IP regions (p = 0.0005).

The highest seroprevalence of 39% (13/33) was noted in Tervetefollowed by 19% (3/16) in Skrunda and 19% (5/27) in Limbazi. Theregional distribution of seropositive dogs in relation to the predom-inant tick species in Latvia is presented in Fig. 1.

86 I. Berzina et al. / Ticks and Tick-borne Diseases 4 (2013) 83– 88

Table 2Age and gender of healthy (HD), healthy hunting (HHD), and sick dogs (SD) tested for A. phagocytophilum antibodies with SNAP 4Dx test.

Parameter HD (n = 400) HHD (n = 41) SD (n = 29)

Seronegative (n = 356) Seropositive (n = 44) Seronegative (n = 36) Seropositive (n = 5) Seronegative (n = 24) Seropositive (n = 5)

Age, yearsMedian 3 5 4 7 6 8Range 0.2–20 1–12 0.6–12 1–11 1–11 2–10

GenderFemale 176 26 18 1 11 1Male 179 18 18 4 13 4Unknown 1 0 0 0 0 0

Table 3Hematology results from the hematology analyzer for healthy and hunting dogs (HD + HHD) and sick dogs (SD). No statistically significant differences (p < 0.05) were notedbetween the seropositive and seronegative dogs in the HD + HHD or in the SD groups.

Variable HD + HHD (n = 421a) SD (n = 29) Reference intervalb

Seronegative (n = 372) Seropositive (n = 49) Seronegative (n = 24) Seropositive (n = 5)

LeukocytesMedian 11.6 12 19 22 6–17 × 109/LRange 4.1–26.6 4.6–22.6 5.4–41.4 18.7–36.1

ErythrocytesMedian 6.8 7 6 7 5.5–8.5 × 1012/LRange 2.3–9.2 2.8–8.2 1.2–8.2 7.1–8.1

PlateletsMedian 275 283 167 104 200–500 × 109/LRange 48–935 81–612 7–585 40–257

re notm ally si

0).

A

wsHtr

tl(

C

aiAlblbb24mttc

P

fn

a Twenty samples from the HD + HHD seronegative dogs contained clots and weicroscopical examination of the blood smears and in the HD + HHD dogs, no clinicb Reference values from Schalm’s Veterinary Hematology 6th ed (Rizzi et al., 201

. phagocytophilum in sick dogs

In the SD group, the seroprevalence was 17% (5/29). Thereas no statistically significant difference between the number of

eropositive dogs in the HD versus the SD group (p = 0.279), nor theHD versus the SD group (p = 0.518). Four dogs with clinical signs

hat were seropositive against A. phagocytophilum were from the IRegion, and one dog was from the M region.

The clinical signs reported for seropositive sick dogs werehrombocytopenia (n = 4), joint problems (n = 3), fever (n = 3),ethargy (n = 3), anorexia (n = 3), and morulae within neutrophilsn = 1).

linical granulocytic anaplasmosis

In one of 5 seropositive dogs, clinical canine granulocyticnaplasmosis was confirmed based on the presence of morulaen neutrophils concurrently with clinical signs compatible with. phagocytophilum infection: fever (rectal temperature 41.2 ◦C),

ethargy, and anorexia. SNAP test showed that this dog had anti-odies against A. phagocytophilum and E. canis; however, the

atter was not confirmed by PCR and was not considered toe associated with the clinical signs. This dog was a mixedreed, female, intact, 8 years old. Hematology results: leukocytes3.2 (6–17 × 109/L), erythrocytes 6.26 (5.5–8.5 × 1012/L), platelets0 (200–500 × 109/L). No platelet aggregates were noted uponicroscopy of the blood smear. Five percent of neutrophils con-

ained one or 2 basophilic round, slightly granular morulae withinhe cytoplasm. No additional clinically significant morphologicalhanges in other leukocytes, erythrocytes, or platelets were noted.

CR

Blood samples from 10 dogs from the SD group were evaluatedor the presence of A. phagocytophilum 16S ribosomal RNA gene byested PCR. This included all 5 seropositive dogs and 5 additional

evaluated by hematology analyzer. However, all analyzer results were verified bygnificant abnormalities were noted.

dogs that were seronegative, but thrombocytopenic. The only posi-tive sample was from the dog with CGA who had intragranulocyticmorulae in the blood smear.

The 16S ribosomal RNA gene sequence of A. phagocytophilumobtained from the dog with CGA was deposited in GenBank withthe accession number JQ966109. It was 100% identical with thesequence of A. phagocytophilum isolated from I. ricinus from Meza-parks in Riga, Latvia (MEGA5 software).

Discussion

Until now, no studies have been published that describe tick-borne diseases in any animal species in Latvia or, to our knowledge,in other Baltic states. This study is the first to evaluate the sero-prevalence of A. phagocytophilum in dogs in Latvia and to determinegeographical distribution of seropositive cases in relation to thedistribution of 2 prevalent tick species in this country. Seropreva-lence was found to be significantly higher in regions where I. ricinusticks prevailed over I. persulcatus. We also described the first caseof clinical canine granulocytic anaplasmosis in Latvia.

No sex or breed disposition for anaplasmosis was seen in ourstudy, and seropositive dogs tended to be older, which is in agree-ment with the previously published data (Carrade et al., 2009;Tsachev, 2009). Our reported seroprevalence in HD (11.0%) andHHD (12%) is comparable with that detected in the geographicallyclosely located countries Sweden (17%) and Poland (21%) (Egenvallet al., 2000b; Engvall and Egenvall, 2002; Welc-Faleciak et al., 2009).Seroprevalence among the HD, HHD, and SD groups in this studywas not statistically different. Similar lack of significant differencewas found by other researchers (Jensen et al., 2007; Kohn et al.,2011).

Serology is a widely used method for determination of expo-

sure to A. phagocytophilum (Beall et al., 2008; Pantchev et al., 2009;Carrade et al., 2011; Kohn et al., 2011). Antibodies develop approx-imately a week after initial exposure and persist for ∼6–8 months.However, they can be decreased below the detection limit after the

ck-bor

tEdedbhedPobpIhobSbP

ehCttbBtmtTa2lbhgsplDife

itewtRwg

c2CAdtRScua

I. Berzina et al. / Ticks and Ti

reatment with antibiotics (Carrade et al., 2011; Kohn et al., 2008;genvall et al., 2000a). Serological test used in this study has beenescribed to be 99% sensitive and 100% specific (Chandrashekart al., 2010; Carrade et al., 2011). Rare serological cross reactivitiesue to exposure to E. canis, Anaplasma platys, and A. marginale haveeen described. However, none of the above-mentioned pathogensave been detected in ticks in Latvia (Bormane, 2007; Carradet al., 2009, 2011; Pantchev et al., 2009). In our SD group, oneog was seropositive for A. phagocytophilum and E. canis. SinceCR was negative for E. canis and confirmed only the presencef A. phagocytophilum, the serological test results of the dog cane explained by cross reactivity between antibodies against A.hagocytophilum in the serum and E. canis antigen in the test.n addition, Rhipicephalus sanguineus ticks that transmit E. canisave not been found in Latvia (Bormane, 2007). Four HD dogs inur study had antibodies against A. phagocytopilum and Borreliaurgdorferi, confirmation by PCR was not performed in these dogs.uch coinfections are reported commonly and are partly explainedy both organisms having a common vector (Beall et al., 2008;antchev et al., 2009).

In our study, a single serological testing was performed. Thus,xposure to A. phagocytophilum in dogs from the SD group mayave been underestimated (Pantchev et al., 2009; Lim et al., 2010;arrade et al., 2011; Rand et al., 2011). It is possible that some ofhe sick dogs in our study were acutely infected with A. phagocy-ophilum, but did not have detectable numbers of morulae in thelood and did not up to then seroconvert (Poitout et al., 2005;eall et al., 2008; Kohn et al., 2011). However, all of the sick dogshat had thrombocytopenia (hematological abnormality most com-

only described in dogs with A. phagocytophilum infection) wereested for A. phagocytophilum by PCR and were found to be negative.he reported detection threshold for nPCR with similar primer pairss used in our study was 0.25 cells/per 0.2 ml (Massung and Slater,003), thus, we can conclude that these dogs at the time of samp-

ing likely were negative for A. phagocytophilum organisms in thelood. Since PCR testing was limited only to the SD group, we mayave missed some persistently infected dogs in the HD and HHDroups that were seronegative. Persistent granulocytic anaplasmo-is is characterized by intermittent rickettsemia and intermittentlyositive PCR and serology without clinical symptoms or hemato-

ogical abnormalities (Egenvall et al., 2000a; Jensen et al., 2007).iagnosis of persistent infection would require a series of serolog-

cal, molecular, and hematological tests that can be carried out inuture studies (Egenvall et al., 2000a; Kohn et al., 2011; Granquistt al., 2010).

Similarly to other authors, we detected no significant differencesn hematological parameters between seropositive and seronega-ive dogs in the HD + HHD and SD groups (Couto et al., 2010; Kohnt al., 2011). Hematology results in the SD group were in agreementith previously published results, with thrombocytopenia being

he most commonly observed abnormality (Lilliehöök et al., 1998;avnik et al., 2011). However, since in our study thrombocytopeniaas one of the selection criteria for the inclusion of dogs in the SD

roup, it could not be used as an independent variable.The clinical anaplasmosis case reported here is typical in its

linical presentation and laboratory abnormalities (Melter et al.,007; Carrade et al., 2009; Granick et al., 2009; Mazepa et al., 2010;anelas Domingos et al., 2011; Ravnik et al., 2011). Our diagnosis of. phagocytophilum infection in this dog was based on the criteriaescribed previously – suggestive clinical signs, morulae in the neu-rophils, seropositivity, and positive PCR test (Carrade et al., 2009;avnik et al., 2011). Acute CGA is a frequent diagnosis in dogs in

weden. Given the roughly similar seroprevalence against A. phago-ytophilum in both countries, we suspect that clinical CGA might benderdiagnosed in dogs in Latvia (Egenvall et al., 2000a,b; Engvallnd Egenvall, 2002).

ne Diseases 4 (2013) 83– 88 87

A. phagocytophilum DNA isolated from the dog with CGA wasidentical to the human prototype strain (GenBank accession num-ber U02521) and to that isolated from I. ricinus ticks in Riga,Latvia (Bormane, 2007). Similarly, A. phagocytophilum strains havebeen isolated from dogs in Germany, Sweden, France, and the USA(Engvall and Egenvall, 2002; Poitout et al., 2005; Canelas Domingoset al., 2011; Silaghi et al., 2011), but an isolate from a dog from Italywas different (Manna et al., 2004). Most of the researchers haveevaluated the variability of A. phagocytophilum strains based on thesequence of 16S rRNA gene (Manna et al., 2004; Kirtz et al., 2005;Silaghi et al., 2011), but analysis of several other genes may providebetter understanding of A. phagocytophilum pathogenicity (Poitoutet al., 2005; Morissette et al., 2009; Canelas Domingos et al., 2011;Kohn et al., 2011; Silaghi et al., 2011).

Previous research on ticks in Latvia has shown distinct regionswhere one or the other tick species prevails and regions whereboth I. ricinus and I. persulcatus ticks are encountered (Bormaneet al., 2004; Bormane, 2007; Karelis et al., 2012). The importance ofregional studies on the incidence of tick-borne diseases has beenstressed by Carrade et al. (2011) as they have found focal areasof high and low seroprevalence of Anaplasma-seropositive dogs inclose proximity to each other in several states in the USA. In Latvia,the highest percentage of seropositive dogs was found around Ter-vete, Skrunda, centrally located I. ricinus regions, and Limbazi, amixed tick region located in the north-central part of Latvia. Ter-vete is rural while the 2 others, Limbazi and Skrunda, are smalltowns. It would be useful to carry out additional studies to evaluateAnaplasma burden in ticks as well as the prevalence of tick-borneinfections in other species of domestic and wild animals in theseregions. Carrade et al. (2009) state that canine granulocytic anaplas-mosis distribution parallels that of the human disease, thus, ourdata could serve as a basis for further investigation on human gran-ulocytic anaplasmosis cases in Latvia.

We conclude that seroprevalence of A. phagocytophilum in dogsin Latvia is related to the distribution of the main vector species,I. ricinus, with seroprevalence being significantly higher in regionswhere I. ricinus is the dominant tick species. Granulocytic anaplas-mosis should be included in the differential list in dogs withappropriate clinical presentation and laboratory abnormalities. A.phagocytophilum seroprevalence in dogs in Latvia is within rangereported for other European countries.

Acknowledgements

This study was funded by European Union and European SocialFunds (Agreement No. 2009/0180/1DP/1.1.2.1.2/09/IPIA/VIAA/017,“Support for Doctoral Studies Program of Latvia University of Agri-culture”, 04.4-08/EF2.D1.20’). SNAP 4Dx tests were kindly donatedby IDEXX laboratories. We thank Ms. Katrine Strode (UniversitatesVetfonds, Latvia) and Mr. Atis Rektins for their technical assistanceand Dr. Liene Dindonis for critical review of the manuscript.

References

Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local align-ment search tool. J. Mol. Biol. 215, 403–410.

Bakken, J.S., Krueth, J., Wilson-Norskog, C., Tilden, R.L., Asanovic, K., Dumler, S.J.,1996. Clinical and laboratory characteristics of human granulocytic ehrlichiosis.JAMA 275, 199–205.

Beall, M.J., Chandrashekar, R., Eberts, M.D., Cyr, K.E., Diniz, P.P., Mainville, C., Hegarty,B.C., Crawford, J.M., Breitschwerdt, E.B., 2008. Serological and molecular preva-lence of Borrelia burgdorferi, Anaplasma phagocytophilum, and Ehrlichia speciesin dogs from Minnesota. Vector Borne Zoonotic Dis. 8, 455–464.

Bormane, A., Lucenko, I., Duks, A., Mavtchoutko, V., Ranka, R., Salmina, K., Baumanis,

V., 2004. Vectors of tick-borne diseases and epidemiological situation in Latviain 1993–2002. Int. J. Med. Microbiol. 293, 36–47.

Bormane, A., 2007. Ixodes ricinus L. and Ixodes persulcatus P.Sch. (Acari: Ixodidae)distribution, significance and molecular epidemiology of transmitted infectiousdiseases in Latvia. Doctoral Thesis. University of Latvia, Riga (in Latvian).

8 ck-bor

C

C

C

C

C

E

E

E

G

G

J

K

K

K

K

K

L

L

L

8 I. Berzina et al. / Ticks and Ti

anelas Domingos, M., Trotta, M., Briend-Marchal, A., Medaille, C., 2011. Anaplas-mosis in two dogs in France and molecular and phylogenetic characterizationof Anaplasma phagocytophilum. Vet. Clin. Pathol. 40, 215–221.

arrade, D.D., Foley, J.E., Borjesson, D.L., Sykes, J.E., 2009. Canine granulocyticanaplasmosis: a review. J. Vet. Intern. Med. 23, 1129–1141.

arrade, D., Foley, J., Sullivan, M., Foley, C.W., Sykes, J.E., 2011. Spatial distributionof seroprevalence for Anaplasma phagocytophilum, Borrelia borgdorferi, Ehrlichiacanis, and Dirofilaria immitis in dogs in Washington, Oregon, and California. Vet.Clin. Pathol. 40, 293–302.

handrashekar, R., Mainville, C.A., Beall, M.J., O’Connor, T., Eberts, M.D., Alleman, A.R.,Gaunt, S.D., Breitschwert, E.B., 2010. Performance of a commercially available in-clinic ELISA for the detection of antibodies against Anaplasma phagocytophilum,Ehrlichia canis, and Borrelia burgdorferi and Dirofilaria immitis antigen in dogs.Am. J. Vet. Res. 71, 1443–1450.

outo, C.G., Lorentzen, L., Beall, M.J., Shields, J., Bertolone, N., Couto, J.I., Couto, K.M.,Nash, S., Slack, J., Kvitko, H., Westendorf, N., Marin, L., Iazbik, C.M., Vicario, F.C.,Sanz, P., Ruano, R., 2010. Serological study of selected vector-borne diseases inshelter dogs in Central Spain using point-of-care assays. Vector-Borne ZoonoticDis. 10, 885–888.

genvall, A., Lilliehöök, I., Bjöersdorff, A., Engvall, E.O., Karlstam, E., Artursson,K., Heldtander, M., Gunnarsson, A., 2000a. Detection of granulocytic Ehrlichiaspecies DNA by PCR in persistently infected dogs. Vet. Rec. 146, 186–190.

genvall, A., Bonnett, B.N., Gunnarsson, A., Hedhammar, Å., Shoukri, M., Bornstein, S.,Artursson, K., 2000b. Seroprevalence of granulocytic Ehrlichia spp. and Borreliaburgdorferi sensu lato in Swedish dogs 1991–94. Scand. J. Infect. Dis. 32, 19–25.

ngvall, E.O., Egenvall, A., 2002. Granulocytic ehrlichiosis in Swedish dogs andhorses. Int. J. Med. Microbiol. 291 (Suppl. 33), 100–103.

ranick, J.L., Armstrong, P.J., Bender, J.B., 2009. Anaplasma phagocytophilum infectionin dogs: 34 cases (2000–2007). J. Am. Vet. Med. Assoc. 234, 1559–1565.

ranquist, E.G., Stuen, S., Crosby, L., Lundgren, A.M., Alleman, A.R., Barbet, A.F., 2010.Variant-specific and diminishing immune responses towards the highly vari-able MSP2 (P44) outer membrane protein of Anaplasma phagocytophilum duringpersistent infection in lambs. Vet. Immunol. Immunopathol. 133, 117–124.

ensen, J., Simon, D., Murua Escobar, H., Soller, J.T., Bullerdiek, J., Beelitz, P., Pfister, K.,Nolte, I., 2007. Anaplasma phagocytophilum in dogs in Germany. Zoonoses PublicHealth 54, 94–101.

arelis, G., Bormane, A., Logina, I., Lucenko, I., Suna, N., Krumina, A., Donaghy, M.,2012. Tick-borne encephalitis in Latvia 1973–2009: epidemiology, clinical fea-tures and sequelae. Eur. J. Neurol. 19, 62–68.

irtz, G., Meli, M., Leidinger, E., Ludwig, P., Thum, D., Czettel, B., Köbl, S., Lutz, H.,2005. Anaplasma phagocytophilum infection in a dog: identifying the causativeagent using PCR. J. Small Anim. Pract. 46, 300–303.

ohn, B., Galke, D., Beelitz, P., Pfister, K., 2008. Clinical features of canine granulocyticanaplasmosis in 18 naturally infected dogs. J. Vet. Intern. Med. 22, 1289–1295.

ohn, B., Silaghi, C., Galke, D., Arndt, G., Pfister, K., 2011. Infections with Anaplasmaphagocytophilum in dogs in Germany. Res. Vet. Sci. 91, 71–76.

ybicova, K., Schanilec, P., Hulinska, D., Uherkova, L., Kurzova, Z., Specjhgalova, S.,2009. Detection of Anaplasma phagocytophilum and Borrelia burgdorferi sensulato in dogs in the Czech Republic. Vector Borne Zoonotic Dis. 9, 655–661.

illiehöök, I., Egenvall, A., Tvedten, H.W., 1998. Hematopathology in dogs experi-mentally infected with a Swedish granulocytic Ehrlichia species. Vet. Clin. Pathol.27, 116–122.

im, S., Irwin, P.J., Lee, S.R., Oh, M.H., Ahh, K.S., Myung, B.Y., Shin, S.S., 2010. Com-

parison of selected canine vector-borne diseases between urban animal shelterand rural hunting dogs in Korea. Parasit. Vectors 3, 32.

iz, J.S., Anderes, L., Sumner, J.W., Massung, R.F., Gern, L., Rutti, B., Brossard, M.,2000. PCR detection of granulocytic Ehrlichiae in Ixodes ricinus ticks and wildsmall mammals in Western Switzerland. J. Clin. Microbiol. 38, 1002–1007.

ne Diseases 4 (2013) 83– 88

Manna, L., Alberti, A., Pavone, L.M., Scibelli, A., Staiano, N., Gravino, A.E., 2004. Firstmolecular characterization of a granulocytic Ehrlichia strain isolated from a dogin South Italy. Vet. J. 167, 224–227.

Massung, R.F., Slater, K.G., 2003. Comparison of PCR assays for detection of theagent of human granulocytic ehrlichiosis, Anaplasma phagocytophilum. J. Clin.Microbiol. 41, 717–722.

Mazepa, A.W., Kidd, L.B., Young, K.M., Trepanier, L.A., 2010. Clinical presentation of26 Anaplasma phagocytophilum-seropositive dogs residing in an endemic area.J. Am. Anim. Hosp. Assoc. 46, 405–412.

Melter, O., Stehlik, I., Kinska, H., Volfova, I., Ticha, V., Hulinska, D., 2007. Infection withAnaplasma phagocytophilum in a young dog: a case report. Vet. Med. (Praha) 52,207–212.

Morissette, E., Massung, R.F., Foley, J.E., Alleman, R.A., Foley, P., Barbet, A.F., 2009.Diversity of Anaplasma phagocytophilum strains, USA. Emerg. Infect. Dis. 15,928–931.

Pantchev, N., Schaper, R., Limousin, S., Norden, N., Weise, M., Lorentzen, L., 2009.Occurrence of Dirofilaria immitis and tick-borne infections caused by Anaplasmaphagocytophilum, Borrelia burgdorferi sensu lato and Ehrlichia canis in domesticdogs in France: results of a countrywide serologic survey. Parasitol. Res. 105,S101–S113.

Parola, P., Raoult, D., 2001. Ticks and tick-borne bacterial diseases in humans: anemerging infectious threat. Clin. Infect. Dis. 32, 897–928.

Paulauskas, A., Radzijevskaja, J., Rosef, O., 2009. Anaplasma in ticks feeding onmigrating birds and questing ticks in Lithuania and Norway. Clin. Microbiol.Infect. 15, 34–36.

Poitout, F.M., Shinozaki, J.K., Stockwell, P.J., Holland, C.J., Shukla, S.K., 2005. Geneticvariants of Anaplasma phagocytophilum infecting dogs in Western WashingtonState. J. Clin. Microbiol. 43, 796–801.

Rand, P.W., Lacombe, E.H., Elias, S.P., Kahill, B.K., Lubelczyk, C.B., Smith, R.P., 2011.Multitarget test for emerging Lyme disease and anaplasmosis in a serosurvey ofdogs, Maine, USA. Emerg. Infect. Dis. 17, 899–902.

Ranka, R., Bormane, A., Salmina, K., Baumanis, V., 2004. Identification of three clin-ically relevant Borrelia burgdorferi sensu lato genospecies by PCR-restrictionfragment length polymorphism analysis of 16S–23S ribosomal DNA spaceramplicons. J. Clin. Microbiol. 42, 1444–1449.

Rar, V.A., Epikhina, T.I., Livanova, N.N., Panov, V.V., Doroschenko, E.K., Pukhovskaya,N.M., Vysochina, N.P., Ivanov, L.I., 2011. Genetic variability of Anaplasma phago-cytophilum in Ixodes persulcatus ticks and small mammals in the Asian part ofRussia. Vector Borne Zoonotic Dis. 11, 1013–1021.

Ravnik, U., Tozon, N., Smrdel, S.K., Zupanc, T.A., 2011. Anaplasmosis in dogs: therelation of hematological, biochemical and clinical alterations to antibody titreand PCR confirmed infection. Vet. Microbiol. 149, 172–176.

Rizzi, T.E., Meinkoth, J.H., Clinkenbeard, D.K., 2010. Normal hematology of the dog.In: Weiss, D.J., Wardrop, K.J. (Eds.), Schalm’s Veterinary Hematology, sixth ed.Wiley-Blackwell, Ames.

Silaghi, C., Kohn, B., Chirek, A., Thiel, C., Nolte, I., Liebisch, G., Pfister, K., 2011.Relationship of molecular and clinical findings on Anaplasma phagocytophiluminvolved in natural infections of dogs. J. Clin. Microbiol. 49, 4413–4414.

Stuen, S., 2007. Anaplasma phagocytophilum – the most widespread tick-borne infec-tion in animals in Europe. Vet. Res. Commun. 31, 79–84.

Süss, J., 2011. Tick-borne encephalitis 2010: epidemiology, risk areas, and virusstrains in Europe and Asia – an overview. Ticks Tick Borne Dis. 2, 2–15.

Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5:

molecular evolutionary genetics analysis using maximum likelihood, evolution-ary distance and maximum parsimony methods. Mol. Biol. Evol. 28, 2731–2739.

Tsachev, I., 2009. Canine granulocytic anaplasmosis. Trakia J. Sci. 7, 68–72.Welc-Faleciak, R., Rodo, A., Sinski, E., Bajer, A., 2009. Babesia canis and other tick-

borne infections in dogs in Central Poland. Vet. Parasitol. 166, 191–198.