Discrimination between six species of canine microfilariae

by a single polymerase chain reaction

Mark Rishniw a,*, Stephen C. Barr b, Kenny W. Simpson b,Marguerite F. Frongillo c, Marc Franz d, Jose Luis Dominguez Alpizar e

a Department of Biomedical Sciences, Box 11, Veterinary Research Tower, College of Veterinary Medicine,

Cornell University, Ithaca, NY 14853, USAb Department of Clinical Sciences, Cornell University, Ithaca, NY, USA

c Department of Microbiology and Immunology, Cornell University, Ithaca, NY, USAd Woodbury Animal Hospital, Woodbury, NY, USA

e Department of Parasitology, Universidad Autonoma de Yucatan, Mexico

Received 16 August 2005; received in revised form 6 October 2005; accepted 6 October 2005

www.elsevier.com/locate/vetpar

Veterinary Parasitology 135 (2006) 303–314

Abstract

Canine dirofilariasis caused by Dirofilaria immitis is usually diagnosed by specific antigen testing and/or identification of

microfilariae. However, D. immitis and at least six other filariae can produce canine microfilaremias with negative heartworm

antigen tests. Discriminating these can be of clinical importance.

To resolve discordant diagnoses by two diagnostic laboratories in an antigen-negative, microfilaremic dog recently imported

into the US from Europe we developed a simple molecular method of identifying different microfilariae, and subsequently

validated our method against six different filariae known to infect dogs by amplifying ribosomal DNA spacer sequences by

polymerase chain reaction using common and species-specific primers, and sequencing the products to confirm the genotype of

the filariae. We identified the filaria in this dog as D. repens. This is the first case of D. repens infection in the United States.

Additionally, we examined microfilariae from five additional antigen-negative, microfilaremic dogs and successfully

identified the infecting parasite in each case. Our diagnoses differed from the initial morphological diagnosis in three of

these cases, demonstrating the inaccuracy of morphological diagnosis. In each case, microfilariae identified morphologically as

A. reconditum were identified as D. immitis by molecular methods. Finally, we demonstrated that our PCR method should

amplify DNA from at least two additional filariae (Onchocerca and Mansonella), suggesting that this method may be suitable for

genotyping all members of the family Onchocercidae.

# 2005 Elsevier B.V. All rights reserved.

Keywords: Dirofilaria; Immitis; Dipetalonema; Acanthocheilonema; Reconditum; Repens; Canine; Knott’s test; Polymerase chain reaction;

PCR

* Corresponding author. Tel.: +1 607 253 3108.

E-mail address: mr89@cornell.edu (M. Rishniw).

0304-4017/$ – see front matter # 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.vetpar.2005.10.013

M. Rishniw et al. / Veterinary Parasitology 135 (2006) 303–314304

1. Introduction

Canine heartworm disease is generally diagnosed

by antigen testing for Dirofilaria immitis, and/or

identification of microfilariae in the blood of infected

dogs. However, other filariae, including Acanthochei-

lonema (Dipetalonema) reconditum, Dirofilaria

(Nochtiella) repens, Acanthocheilonema (Dipetalo-

nema) dracunculoides, Cercopithifilaria grassi, Bru-

gia malayi, Brugia pahangi, Brugia ceylonensis and

approximately 1% of Dirofilaria immitis infestations,

can produce persistent microfilaremias with negative

heartworm antigen tests (Courtney et al., 1993;

Fischer et al., 2002; Genchi, 2003). In such cases,

diagnosis relies on examination of the microfilariae,

generally using a modified Knott’s test, or alkaline

phosphatase staining (Chalifoux and Hunt, 1971;

Knott, 1935). These methods are imperfect, and rely

on specialist training to accurately differentiate the

filariae. However, in regions where one of these

filariae is not enzootic, parasitologists may exclude

that filaria from consideration, or fail to recognize it,

especially if a history of potential exposure is not

provided (Cordonnier et al., 2002).

Accurate identification of filarial species in dogs

can be clinically important, because of the zoonotic

concerns and therapeutic implications. While A.

reconditum and A. dracunculoides have few (if any)

clinical consequences, D. repens infestations have

been associated with subcutaneous granulomas and

pruritus in dogs (Baneth et al., 2002; Bredal et al.,

1998; Manuali et al., 2005; Tarello, 2003), and

subcutaneous, conjunctival and pulmonary nodules,

often confused with neoplastic tumors, in humans

(Cordonnier et al., 2002; Pampiglione et al., 1996,

2001; Romano et al., 1976; Vakalis and Himonas,

1997). To reduce the risk of zoonotic transmission,

treatment of D. repens infestations with filaricidal

agents, or at least microfilaricidal agents, is recom-

mended (adulticidal therapy can be postponed until

clinical signs occur if risk of treatment is deemed to

outweigh risk of disease). Adult D. repens have been

successfully treated with melarsomine hydrochloride

(Immiticide, Merial; two injections of 2.5 mg/kg IM

24 h apart) (Baneth et al., 2002). Microfilariae are

susceptible to microfilaricidal doses of avermectins

(ivermectin 50 mg/kg PO or SQ; doramectin 400 mg/

kg SQ; effects of milbemycin have not been

investigated) (Baneth et al., 2002; Tarello, 2003).

On the other hand, treatment of A. dracunculoides or

A. reconditum infestations is not usually required, so

they should be positively identified to avoid inap-

propriate adulticide therapy.

Over the last decade, several investigators have

examined the molecular differentiation of filariae,

using Southern blotting to identify DNA repeats from

D. repens (Chandrasekharan et al., 1994) or poly-

merase chain reaction (PCR) to differentiate either D.

repens or A. reconditum from D. immitis (Bredal et al.,

1998; Favia et al., 1996, 1997; Mar et al., 2002;

Vakalis et al., 1999). However, no existing studies

examined the ability of PCR to differentiate between

more than two filariae. While it would be possible to

run multiple PCR reactions using published primer

sequences to identify a specific filaria, a single

differentiating reaction would provide an expeditious

means of obtaining a diagnosis. Additionally,

sequences are not available for all filariae of interest

at this time, so species-specific reactions may fail in

these cases.

This study was prompted by several clinical cases

requiring accurate identification of microfilariae in

dogs. These dogs had negative heartworm antigen

tests and, in several instances, discrepant morpholo-

gical identifications of the microfilariae. Using

molecular techniques, we sought to accurately identify

the filarial species in each case, and subsequently

developed a PCR-based method that distinguishes

between the six species tested to date (D. immitis, A.

reconditum, D. repens, A. dracunculoides, B. pahangi,

B. malayi) with a single primer pair. Our technique

simplifies filarial identification, compared with pre-

vious studies.

2. Materials and methods

2.1. Case reports

2.1.1. Case 1

A 2-year-old intact male Labrador Retriever, born

in the Czech Republic, but exported as a puppy to The

Netherlands, was presented to the referring veterinar-

ian (MF) for routine evaluation after importation from

The Netherlands in January 2004 to Canada and

subsequently New York State (United States) in April

M. Rishniw et al. / Veterinary Parasitology 135 (2006) 303–314 305

2004. The dog was clinically normal and had never

been tested for heartworm or administered heartworm

preventative. The veterinarian identified a microfilar-

emia on routine heartworm screening, and performed

a heartworm antigen test using a Dirocheck ELISA

(Synbiotics Corporation, San Diego, CA), which was

negative. Repeated antigen testing at several diag-

nostic laboratories failed to diagnose D. immitis

infestation. Consequently, the veterinarian submitted

blood for morphological classification of microfilariae

to two diagnostic laboratories. Examination of the

microfilariae using modified Knott’s tests resulted in

discrepant diagnoses: one laboratory identified the

microfilariae as A. reconditum, while the other

identified them as D. repens.

2.1.2. Case 2

An 11-year-old male neutered mixed breed dog

was presented to a referring veterinarian in San Diego,

California, for routine evaluation. As with the first

case, a microfilaremia was detected, but antigen

testing for D. immitis was negative on multiple

occasions. The microfilariae were identified as A.

reconditum by a commercial diagnostic laboratory.

2.1.3. Case 3

An 8-year-old dog (signalment unknown) was

presented to the referring veterinarian in Utica, New

York, for evaluation of hematuria secondary to

prostatitis. Microfilariae were detected on urinalysis,

and subsequently on examination of blood. Antigen

tests for D. immitis were negative. The microfilariae

were morphologically identified as D. immitis with a

Knott’s test.

2.1.4. Case 4

An 8-year-old female speyed miniature schnauzer

was presented to the referring veterinarian after

adoption from an animal shelter in Clemmons, North

Carolina. Again, a microfilaremia was detected and

identified morphologically as A. reconditum, but

antigen testing for D. immitis was negative on

multiple occasions.

2.1.5. Case 5

A 1-year-old male intact Pug was presented for

heartworm testing in Modesto, California. Microfilar-

emia was detected and identified morphologically as

A. reconditum, but antigen testing for D. immitis was

negative.

2.1.6. Case 6

An 8-year-old male neutered Malamute was

presented for heartworm testing in Vallejo, California.

Microfilaremia was detected, and identified morpho-

logically as A. reconditum but antigen testing for D.

immitis was negative.

2.2. Sample preparation

In all six clinical cases, 1 ml of blood (antic-

oagulated with heparin or EDTA) was submitted.

We used blood from a microfilaremic dog with D.

immitis infestation and from a dog with A. reconditum

infestation as positive controls. In both controls

antigen testing and Knott’s testing provided the

filarial diagnosis. The sample of A. reconditum-

infested blood was obtained by one of the authors and

morphologically identified as A. reconditum. Fresh

adult B. malayi and B. pahangi worms were

generously provided by T. Klei of Louisiana State

University. Blood from a dog infested with A.

dracunculoides was generously provided by Francisco

Alonso de Vega of Murcia University in Spain after

morphologic diagnosis with alkaline phosphatase

staining.

Finally, we used DNA from a non-infected dog as a

negative control in all PCR reactions.

2.3. DNA extraction

In all clinical cases, the D. immitis positive control

specimen and the Brugia specimens, 500 ml antic-

oagulated blood (or five adult Brugia filariae) were

suspended in 500 ml lysis buffer (50 mM Tris, pH 8.0;

100 mM EDTA, 100 mM NaCl, 1% SDS) and

digested with 10 ml proteinase K (10 mg/ml) for

16 h at 55 8C, followed by a standard phenol/chloro-

form extraction method, as described previously

(Sambrook and Russell, 2001).

The A. reconditum and A. dracunculoides positive

control specimens were submitted on filter paper as

dried blood spots. DNAwas extracted using a QIAmp

DNA Blood Mini Kit DNA extraction kit (Qiagen

Inc., Valencia, CA) according to manufacturer’s

instructions.

M. Rishniw et al. / Veterinary Parasitology 135 (2006) 303–314306

2.4. Polymerase chain reaction filarial genotyping

We designed primer pairs that spanned the internal

transcribed spacer region 2 (ITS2) of the ribosomal

DNA of either D. immitis, or A. reconditum, and a

primer pair spanning the 5S ribosomal intergenic region

of D. repens based on previously published sequences

(Table 1) (Favia et al., 2000; Mar et al., 2002). These

primers were used to specifically genotype each filarial

species. Additionally, we designed a primer pair that

would amplify fragments of different fragment length

from both D. immitis and A. reconditum (DIDR-F1,

DIDR-R1) (referred to as the pan-filarial primers

throughout the remainder of the manuscript).

For additional independent genotyping validation,

we designed primers from published sequences of the

cytochrome oxidase subunit 1 (COI) gene for each of

these species (Table 1).

Each PCR reaction consisted of 1.5 mM MgCl2,

250 mM dNTPs, 20 mM Tris–HCl (pH 8.4), 50 mM

Table 1

Primer sequences used to amplify PCR products from filarial and canine

Primer pair Primer sequence Gene ta

D.imm-F1a CAT CAG GTG ATG ATG TGA TGA T ITS2

D.imm-R1 TTG ATT GGA TTT TAA CGT ATC ATT T

A.rec-F1a CAG GTG ATG GTT TGA TGT GC ITS2

A.rec-R1 CAC TCG CAC TGC TTC ACT TC

D.rep-F1 TGT TTC GGC CTA GTG TTT CGA CCA 5SrRNA

D.rep-R1 ACG AGA TGT CGT GCT TTC AAC GTG

DIDR-F1 AGT GCG AAT TGC AGA CGC ATT GAG 5.8S-IT

DIDR-R1 AGC GGG TAA TCA CGA CTG AGT TGA

DI COI -F1 AGT GTA GAG GGT CAG CCT GAG TTA COI

DI COI-R1 ACA GGC ACT GAC AAT ACC AAT

AR COI-F1 AGT GTT GAG GGA CAG CCA GAA TTG COI

AR COI-R1 CCA AAA CTG GAA CAG ACA AAA CAA GC

DR COI-F1 AGT GTT GAT GGT CAA CCT GAA TTA COI

DR COI-R1 GCC AAA ACA GGA ACA GAT AAA ACT

Actin F2 ACC ACT GGT ATT GTC ATG GAC TCT G Canine

Actin R2 GCT CTT CTC CAG GGA GGA CGA

a D.imm-F1, D.imm-R1, A.rec-F1 and A.rec-R1 are identical to ITS2F-D

(2002).b Predicted product size based on Genbank sequence.

KCl, 2.5 U of Taq polymerase and 1 ml of DNA

solution in a total volume of 20 ml. The PCR

procedure consisted of a denaturing step at 94 8Cfor 2 min and 32 cycles of denaturing (30 s at 94 8C),annealing (30 s at 60 8C for ITS2-based primers and

30 s at 58 8C for COI-based primers) and extension

(30 s at 72 8C); a final extension (7 min at 72 8C) and asoak at 4 8C in an Eppendorf Mastercycler thermal

cycler (Brinkmann Instruments Inc., Westbury, NY).

The PCR product (10 ml) was examined on a 1.5%

agarose gel, and fragments of predicted size were

extracted from both reactions and purified using a

QiaQuick PCR purification kit (Qiagen Inc., Valencia,

CA). The PCR products were ligated into pGEM-T

Easy Vector System 1 TA cloning vectors (Promega

US, Madison, WI) and cloned (Sambrook and Russell,

2001). Candidate clones were screened by restriction

digests, and clones with expected digestion fragments

were sequenced at the institutional sequencing facility

on an Applied Biosystems Automated 3730 DNA

DNA

rget Primer size

(base-pairs)

Product

origin

Product size

(base-pairs)

Genbank

accession #

22 D. immitis 302 AF217800

25

20 A. reconditum 348 AF217801

20

24 D. repens 247 and 153 AJ242966

24 AJ242967

S2-28S 24 D. immitis 542 AF217800

24 A. reconditum 578 AF217801

D. repens 484 AY693808

A. dracunculoides 584 DQ018785

B. pahangi 664 AY988600

B. malayi 615 AY988599

B. timori 625b AF499130

M. ozzardi 430b AF228554

O. volvulus 470b AF228575

24 D. immitis 203 AB192887

21

24 A. reconditum 208 AJ544876

26

24 D. repens 209 AJ271614

24

b-actin 25 Canis familiaris 276 AF021873

21

i, ITS2R-Di, ITS2F-Dr and ITSR-Dr primers published byMar et al.

M. Rishniw et al. / Veterinary Parasitology 135 (2006) 303–314 307

Fig. 1. Gel electrophoresis of filarial PCR products in a 1.5%

agarose gel using filarial-specific ITS2-region primers. Lanes 1,

6, 11, 16: MW marker (KB Plus); lanes 2, 7, 12: D. immitis DNA;

lanes 3, 8, 13: A. reconditum DNA; lanes 4, 9, 14: D. repens (Case 1)

DNA; lanes 5, 10, 15: C. familiaris DNA (negative control). Lanes

2–5: D.immF1/D.immR1 primer pair; lanes 7–10: A.recF1/A.recR1

primer pair; lanes 12–15: D.repF1/D.repR1 primer pair.

Analyzer (Applied Biosystems, Foster City, CA) using

universal M13 forward and reverse primers. Sequence

results were compared with those of other species

within the Genbank data bank at the National Center

for Biomedical Information.

Additionally, we examined DNA quality by

performing PCR for canine actin on each sample,

to ensure sufficient template for the microfilarial

PCR.

2.5. Microfilarial phenotyping

In addition to the PCR genotyping, we performed a

modified Knott’s test and wet smear on the blood

submitted from Cases 1–3 and the positive control

samples. The morphological diagnosis for the six

clinical cases is the preliminary morphological

diagnosis provided with initial sample submission

and not the diagnosis obtained by one of the authors

(MFF).

Fig. 2. Gel electrophoresis of filarial PCR products in a 1.5%

agarose gel using pan-filarial ITS2-region primers. Lanes 1, 4:

MW marker (KB Plus); lane 2: A. reconditum DNA; lane 3: D.

immitis DNA; lane 3: D. repens (Case 1) DNA.

3. Results

All blood-derived DNA samples were positive for

canine actin (not shown), indicating adequate DNA

extraction.

Primers specific for D. immitis (D.imm-F1 and

D.imm-R1), A. reconditum (A.rec-F1 and A.rec-R1)

and D. repens (D.rep-F1 and D.rep-R1) amplified the

expected products only from D. immitis (302 bp), A.

reconditum (348 bp) and D. repens (247 bp, 153 bp)

DNA, respectively (Fig. 1). The doublet amplicon

obtained with primers specific for D. repens has been

previously reported (Favia et al., 2000). There was

marginal amplification of a product from A. recondi-

tum DNA (247 bp) with D. repens-specific primers.

No filariae-specific primer pairs amplified any

products from DNA of the normal dog (negative

control), or from DNA extracted from B. pahangi,

B. malayi or A. dracunculoides (data not shown).

These results confirmed the genotyping of D. immitis,

A. reconditum and D. repens for subsequent analysis

and the specificity of the primer sets.

The pan-filarial primer pair (DIDR-F1 and DIDR-

R1) amplified the expected product from D. immitis

DNA (542 bp), and A. reconditum DNA (578 bp)

(Fig. 2). Unexpectedly, this primer pair also amplified

a 484 base-pair product from the DNA extracted from

Case 1 (D. repens-infested dog). The different sizes of

the amplicons for all three filarial species allowed us to

use this pan-filarial primer set to genotype the filarial

species in Cases 2–6.

Filaria-specific primers for COI demonstrated

specificity for each of the filarial species (Fig. 3),

further confirming the genotyping of microfilariae

from Case 1 as D. repens and validating the pan-

filarial primer set results.

Results of PCR with pan-filarial primers from

Cases 2–4 and 6 identified the microfilariae as D.

immitis (Fig. 4) (confirmed with species-specific COI

PCR); PCR from Case 5 identified the microfilariae as

A. reconditum (also confirmed with species-specific

COI PCR).

M. Rishniw et al. / Veterinary Parasitology 135 (2006) 303–314308

Fig. 3. Gel electrophoresis of filarial PCR products in a 1.5%

agarose gel using filarial-specific COI primers. Lanes 1, 5, 9, 13:

MW marker (KB Plus); lanes 2, 6, 10: D. immitis DNA; lanes 3, 7,

11: A. reconditum DNA; lanes 4, 8, 12: D. repens (Case 1) DNA.

Sequencing of the two amplicons obtained with the

D. repens-specific ITS2-region primers (D.rep-F1,

D.rep-R1) unequivocally confirmed the presumptive

diagnosis of D. repens microfilaremia in Case 1.

Sequencing of the 484 bp amplicon obtained with the

pan-filarial primers identified a novel sequence that

Fig. 4. Gel electrophoresis of filarial PCR products in a 1.5%

agarose gel using pan-filarial ITS2-region primers. Lanes 1, 6:

MW marker (KB Plus); lane 2: Case 4 DNA; lane 3: D. immitis

DNA; lane 4: A. reconditum DNA; lane 5: D. repens (Case 1) DNA.

had high homology with the 5.8S and 28S ribosomal

RNA sequences from D. immitis (present at the 50 and30 termini of the amplicon), but only moderate

homology with the central D. immitis ITS2 region

(Fig. 5), and no appreciable homology with A.

reconditum ITS2 region. This sequence was conse-

quently labeled as the ITS2 region of D. repens,

submitted to the Genbank data bank at the National

Center for Biomedical Information and assigned the

accession number AY693808.

Morphologic examination of the microfilariae from

Case 1 using a modified Knott’s technique, showed a

lack of a cephalic hook, a slightly tapering anterior

end, and no consistent tail bend (Fig. 6). The

concentration of microfilariae (abundant) and the

motility of live microfilariae (not propulsive) on a wet

smear, were most consistent with a diagnosis of D.

repens. Examination of the microfilariae from Cases 2

and 3 by one of the authors (MFF) supported our

molecular diagnosis of D. immitis.

Results of PCR from B. malayi, B. pahangi and A.

dracunculoides DNA, using the pan-filarial primers,

produced bands at 615 bp (B. malayi), 664 bp (B.

pahangi) and 584 bp (A. dracunculoides) (data not

shown). Sequencing of these amplicons revealed

unique sequences, which were subsequently submitted

to Genbank and assigned accession numbers

AY988599 (B. malayi), AY988600 (B. pahangi) and

DQ018785 (A. dracunculoides).

While PCR with pan-filarial primers failed to

discriminate between A. reconditum and A. dracun-

culoides, restriction fragment length polymorphism

PCR (RFLP-PCR) using MseII resulted in two

fragments (251 bp and 333 bp) only in A. dracuncu-

loides (data not shown) allowing discrimination

between these two Acanthocheilonema species.

Sequence comparison of the ITS2 region amplified

by our pan-filarial primers (DIDR-F1 and DIDR-R1)

and previously published filarial sequences of this

region demonstrated 77–86% similarity between

species of the same genus (e.g. D. immitis and D.

repens showed 77% similarity, all three Brugia species

showed 86% similarity), but only 30–60% similarity

between species of different genera (e.g. B. pahangi

and A. reconditum showed only 30% similarity).

Phylogenetic tree analysis also grouped sequences

most closely by genus, i.e., species of the same genus

grouped together.

M.

Rish

niw

eta

l./Veterin

ary

Pa

rasito

log

y1

35

(20

06

)3

03

–3

14

309

Fig. 5. Nucleotide sequence alignment of 5.8S-ITS2-28S ribosomal DNA region of seven filariae. The 5.8S and 28S ribosomal DNA are identified above the 50 and 30 ends of thesequences; shaded sequences within the majority sequence are primer binding sequences for the pan-filarial primers.

M. Rishniw et al. / Veterinary Parasitology 135 (2006) 303–314310

Fig.5.(C

on

tin

ued

).

M. Rishniw et al. / Veterinary Parasitology 135 (2006) 303–314 311

Fig. 6. Modified Knott’s stain of microfilaria from Case 1 (D. repens). The head is slightly tapered, there is no cephalic hook, and the tail is

straight, consistent with Dirofilaria repens. Magnification 100�.

4. Discussion

We initiated this study to resolve a discordant

diagnosis of filariasis in a dog, and successfully

identified a D. repens infestation in a dog in the United

States by PCR genotyping of microfilarial DNA

extracted from the patient’s blood. We also resolved

discordant diagnoses provided by standard morpholo-

gical examination of the microfilariae (Knott’s test).

Unexpectedly, the pan-filarial primers we designed

(DIDR-F1 and DIDR-R1), based on sequence homol-

ogy ofD. immitis andA. reconditum, amplified not only

the expected 542 and 578 bp products from these

filariae, but also a 484 bp product from D. repens, that

was present in the Case 1 of this study, a 584 bp product

from A. dracunculoides, a 615 bp product from B.

malayi, and a 664 bp product from B. pahangi. This

permitted a single PCR reaction to differentiate

between D. immitis and five other filariae found in

dogs, removing the need to perform multiple species-

specific genotyping reactions. We subsequently identi-

fied microfilariae from an additional four dogs as D.

immitis and as A. reconditum in one dog with this pan-

filarial primer set. Our molecular identification was

concordant with independent morphological phenotyp-

ing of themicrofilariae by an experienced parasitologist

(MFF), but discordant with the initial morphological

diagnosis in 3/5 cases, illustrating the inaccuracy of

morphological diagnosis by minimally trained techni-

cians. Finally, we were able to provide additional

genotyping information for several filariae that usually

require specialized diagnostic interpretation.

Previous investigators have utilized PCR technol-

ogy to diagnose filarial infestations in dogs and

humans (Baneth et al., 2002; Favia et al., 1996, 1997;

Mar et al., 2002); however, in each of these studies,

only two filarial species were compared. We designed

our pan-filarial primers from previously submitted

sequences, and optimized them for efficient amplifi-

cation.

The need to correctly identify the filarial species is

becoming relevant throughout the world, because of

the increased frequency of transportation of dogs

between continents and countries (Genchi, 2003;

Zahler et al., 1997). While D. repens has been

considered an exotic parasite in some parts of the

world, such as the United States and Australia, and

consequently not considered in the differential

diagnosis of canine filariasis in these continents, this

case illustrates the potential diagnostic complications

that can arise, and provides a simple molecular

strategy for correctly identifying the filaroid. The

potential mosquito vectors of D. repens—Aedes

aegypti (Cancrini et al., 2003), Ae. albopictus (Any-

anwu et al., 2000) and Ae. vexans—are enzootic in the

United States and Ae. vexans is enzootic in the region

of New York State where this dog currently lives

(Nassau County Department of Health, 2003).

Similarly, Ae. aegypti is enzootic in Australia; Ae.

albopictus is not found in Australia. Thus, the

potential exists for spread of D. repens in these

regions. Whether the strains in the United States or

Australia can incubate larvae and transmit the

infestations is unknown. Interestingly, Ae. aegypti is

M. Rishniw et al. / Veterinary Parasitology 135 (2006) 303–314312

relatively resistant to infestation with D. immitis, and

is considered an inefficient intermediate host for

transmission of heartworm disease (Apperson et al.,

1989).

The geographic region in which Case 1 acquired

the infection is unknown. Dirofilaria repens has not

been reported in The Netherlands, but has been

identified in the Czech Republic (where this dog was

born), Hungary, Italy, Greece and France (Cordonnier

et al., 2002; Elek et al., 2000; Pampiglione et al., 2001;

Vakalis and Himonas, 1997; Vasilkova et al., 1992).

The microfilariae from Case 2 were initially

incorrectly identified as A. reconditum, partly based

on a negative antigen test, while the microfilariae in

Case 3 had been identified correctly as D. immitis. In

Cases 2 and 3, examination by an experienced

parasitologist (MFF) correctly identified the micro-

filariae as D. immitis. Causes for antigen-negative

D. immitis microfilaremias include prior adulticide

therapy without subsequent microfilaricidal therapy or

natural death of adult heartworm with persistence of

microfilariae (which have a lifespan of approximately

2 years) (Otto, 1977). Additionally, at least one

product insert (Dirocheck Canine Heartworm Antigen

Test Kit Product Insert, Synbiotics, San Diego, CA)

also lists subthreshold antigenemia, immune clearance

of antigen–antibody complexes, false positive micro-

filarial results (due to contamination of reagents used

for sample preparation), transfusions from micro-

filaremic dogs, or destruction of antigen due to

improper storage as potential causes of discordant

results. While a Dirofilaria microfilaremia does not

necessarily endanger the health of the infested patient,

the patient can act as a reservoir for further infestation

of other dogs if the intermediate mosquito host is

present. Microfilaricidal therapy can resolve this

situation.

The microfilaremia in Case 3 was initially

identified during a routine urinalysis for hematuria

(microfilaruria). We believe this is the first reported

case of microfilaruria in a dog. A prior report exists of

microfilaruria in a cat (Beaufils et al., 1991).

We did not test the discriminatory ability of our

pan-filarial primer set against all filariae that are

known to infest dogs or cats (Genchi, 2003). At least

one additional ‘‘Dipetalonema-like’’ filaria exists in

Europe and Africa—Cercopithifilaria (Acanthochei-

lonema) grassi—but has not been reported in the

United States. It is possible that our pan-filarial primer

set will fail to discriminate this filaria from D. immitis,

however, D. immitis-specific primer sets should

correctly genotype the filarial species in question if

this becomes necessary. Further studies are required to

validate the pan-filarial primers against C. grassi.

Additionally, reports exist of novel filariae in dogs in

Ireland and the United States (Jackson, 1975; Pacheco

and Tulloch, 1970). Techniques such as those

described in this report might be useful in genotyping

such filariae.

The B. malayi sequence we obtained from adult

filariae showed only 86% similarity with a published

B. malayi sequence (Genbank # AF499130), compar-

able to the degree of similarity between our B. pahangi

sequence and the published B. malayi sequence

(AF499130), or between our B pahangi and B. malayi

sequences. This differed substantially from the

similarity between the published sequences for B.

malayi (AF499130) and B. timori (AF499132), which

the authors claimed were 99.5% similar, with only 4

base-pair substitutions along the 600 base-pair region

(Fischer et al., 2002). Several reasons exist for these

discrepant results. Fischer et al claimed that this

degree of similarity was unique to these two species,

however, they did not examine DNA from adult

Brugia species, but only from microfilaria, and by

supposition based on epidemiological data. Thus, their

initial phenotypic classification, which they used as a

gold standard for that study, could be erroneous. We

could find no study by these authors that confirmed

their results using adult B. malayi and B. timori. Given

the much lower degree of similarity between species

of a single genus detailed in this report (e.g. D. immitis

and D. repens, B. malayi and B. pahangi, or A.

reconditum and A. dracunculoides all showed intra-

genus, interspecies variation of approximately 20%,

similar to the difference we observed between B.

malayi and the previously published B. timori), we

suspect that the sequence for B. malayi (AF499130)

published by Fischer et al. (2002) is incorrect and is in

fact the sequence for B timori. Examination of the

data presented by these authors indicates a broad

genetic variation of the Hha1 tandem repeat in these

species, which also suggests a misclassification of

microfilariae.

A less plausible alternative is that inter-species

variation of the ITS2 region between B. malayi and B.

M. Rishniw et al. / Veterinary Parasitology 135 (2006) 303–314 313

timori (<0.5%) is less than the intra-species variation

of B. malayi from different regions (>20%). Several

studies of ITS2 regions in various helminths suggest

that while intra-species variation can occur, it is less

than inter-species variation, and that in most cases

intra-species variation is limited to single nucleotide

polymorphisms (Conole et al., 1999; Gasser et al.,

1999a,b; Huby-Chilton et al., 2001; Newton et al.,

1998;Woods et al., 2000) and previous studies of other

filariae show similar results (Gasser et al., 1996).

These findings agree with our observations of ITS2

variability in filariae and support our genotypic

identification of B. malayi. Finally, it is possible that

the sample of adult B. malayi provided for our study

was incorrectly identified; however, phenotypic

classification of filariae is generally more robust

using adult filariae than microfilariae. The reason for

the discrepancy between our data and that of Fischer

et al. (2002) remains to be elucidated.

Finally, we examined the ability of our pan-filarial

primer sequences to amplify the 5.8S-ITS2-28S

region from two additional filariae in silico. Analysis

of published sequences for Onchocerca volvulus

(Genbank # AF228575), and Mansonella ozzardi

(Genbank # AF228554) demonstrated that our pan-

filarial primers should amplify 470 and 430 bp

amplicons, respectively. This suggests that our primer

set might be uniquely capable of amplifying DNA that

can be used to genotype all filariae of the family

Onchocercidae.

We recommend using the pan-filarial ITS2-region

DIDR-F1 and DIDR-R1 primers with DNA from

D. immitis (as a positive control and amplicon size

standard) against which to compare amplicons from

clinical cases of microfilaremia that test negative for

D. immitis antigen. Amplicons approximately 40 base-

pairs larger than the D. immitis amplicon (542 base-

pairs) are diagnosed as A. reconditum (578 base-pairs)

or A. dracunculoides (584 base-pairs), while those

approximately 60 base-pairs smaller than D. immitis

are diagnosed as D. repens (484 base-pairs). Dis-

crimination of Brugia species by comparing ampli-

cons is limited because B. malayi (615 base-pairs) is

similar to B. timori (625 base-pairs), but these species

should be easily differentiated from B. pahangi (664

base-pairs). However, PCR-RFLP will discriminate B.

malayi and B. timori. If further confirmation is req-

uired, either species-specific PCR can be performed,

using the published primer pairs from this and other

studies, or the amplicons can be sequenced, however,

we feel that simple gel electrophoresis should be

sufficient for diagnosis in most cases. Additionally, we

feel that clinicians and parasitologists should be

alerted to the possibility that D. repens infestation

might be an emerging condition with zoonotic

potential in the United States.

Acknowledgements

The authors would like to thank Dwight Bowman,

Susan Wade and Stephanie Schaaf for their technical

insights and providing D. immitis positive control

samples; Andrew Fei and Fabiano Montiani Ferreira

for assistance in obtaining A. reconditum samples;

Thom Klei for generously providing Brugia filaria;

Francisco Alonso de Vega for generously providing

A. dracunculoides samples and Veterinary Informa-

tion Network for providing access to the case that

initiated this study.

References

Anyanwu, I.N., Agbede, R.I., Ajanusi, O.J., Umoh, J.U., Ibrahim,

N.D., 2000. The incrimination of Aedes (stegomyia) aegypti as

the vector of Dirofilaria repens in Nigeria. Vet. Parasitol. 92,

319–327.

Apperson, C.S., Engber, B., Levine, J.F., 1989. Relative suitability

of Aedes albopictus and Aedes aegypti in North Carolina to

support development of Dirofilaria immitis. J. Am. Mosq.

Control Assoc. 5, 377–382.

Baneth, G., Volansky, Z., Anug, Y., Favia, G., Bain, O., Goldstein,

R.E., Harrus, S., 2002. Dirofilaria repens infection in a dog:

diagnosis and treatment with melarsomine and doramectin. Vet.

Parasitol. 105, 173–178.

Beaufils, J.P., Martin-Granel, J., Bertrand, F., 1991. Presance de

microfilaires de Dirofilaria immitis dans les urines d’un chat

occlus. Prat. Med. Chir. An. Com. 26, 467–472 in French.

Bredal, W.P., Gjerde, B., Eberhard, M.L., Aleksandersen, M.,

Wilhelmsen, D.K., Mansfield, L.S., 1998. Adult Dirofilaria

repens in a subcutaneous granuloma on the chest of a dog. J.

Small Anim. Pract. 39, 595–597.

Cancrini, G., Romi, R., Gabrielli, S., Toma, L., Di Paolo, M.,

Scaramozzino, P., 2003. First finding of Dirofilaria repens in

a natural population of Aedes albopictus. Med. Vet. Entomol. 17,

448–451.

Chalifoux, L., Hunt, R.D., 1971. Histochemical differentiation of

Dirofilaria immitis and Dipetalonema reconditum. J. Am. Vet.

Med. Assoc. 158, 601–605.

M. Rishniw et al. / Veterinary Parasitology 135 (2006) 303–314314

Chandrasekharan, N.V., Karunanayake, E.H., Franzen, L., Abeye-

wickreme, W., Pettersson, U., 1994. Dirofilaria repens: cloning

and characterization of a repeated DNA sequence for the diag-

nosis of dirofilariasis in dogs, Canis familiaris. Exp. Parasitol.

78, 279–286.

Conole, J.C., Chilton, N.B., Jarvis, T., Gasser, R.B., 1999. Intras-

pecific and interspecific variation in the second internal tran-

scribed spacer (ITS-2) sequence for Metastrongylus (Nematoda:

Metastrongyloidea) detected by high resolution PCR-RFLP. Int.

J. Parasitol. 29, 1935–1940.

Cordonnier, C., Chatelain, D., Nevez, G., Sevestre, H., Gontier,

M.F., Raccurt, C.P., 2002. Difficulties in diagnosis of human

dirofilariasis outside known enzootic areas. Rev. Med. Interne

23, 71–76.

Courtney, C.H., Zeng, Q.Y., MacKinnon, B.R., 1993. Comparison of

two antigen tests and the modified Knott’s test for detection of

canine heartworm at different worm burdens. Canine Pract. 18,

5–7.

Elek, G., Minik, K., Pajor, L., Parlagi, G., Varga, I., Vetesi, F.,

Zombori, J., 2000. New human Dirofilarioses in Hungary.

Pathol. Oncol. Res. 6, 141–145.

Favia, G., Lanfrancotti, A., Della Torre, A., Cancrini, G., Coluzzi,

M., 1996. Polymerase chain reaction-identification of Dirofi-

laria repens and Dirofilaria immitis. Parasitology 113 (Pt. 6),

567–571.

Favia, G., Lanfrancotti, A., della Torre, A., Cancrini, G., Coluzzi,

M., 1997. Advances in the identification of Dirofilaria repens

and Dirofilaria immitis by a PCR-based approach. Parassitologia

39, 401–402.

Favia, G., Bazzocchi, C., Cancrini, G., Genchi, C., Bandi, C., 2000.

Unusual organization of the 5S ribosomal spacer in Dirofilaria

repens: absence of a canonical spliced leader 1 sequence.

Parasitol. Res. 86, 497–499.

Fischer, P., Wibowo, H., Pischke, S., Ruckert, P., Liebau, E., Ismid,

I.S., Supali, T., 2002. PCR-based detection and identification of

the filarial parasite Brugia timori from Alor Island, Indonesia.

Ann. Trop. Med. Parasitol. 96, 809–821.

Gasser, R.B., LeGoff, L., Petit, G., Bain, O., 1996. Rapid delineation

of closely-related filarial parasites using genetic markers in

spacer rDNA. Acta Trop. 62, 143–150.

Gasser, R.B., Rossi, L., Zhu, X., 1999a. Identification ofNematodirus

species (Nematoda: Molineidae) from wild ruminants in Italy

using ribosomal DNA markers. Int. J. Parasitol. 29, 1809–1817.

Gasser, R.B., Woods, W.G., Huffman, M.A., Blotkamp, J., Polder-

man, A.M., 1999b. Molecular separation of Oesophagostomum

stephanostomum and Oesophagostomum bifurcum (Nematoda:

Strongyloidea) from non-human primates. Int. J. Parasitol. 29,

1087–1091.

Genchi, C., 2003. Epidemiology and distribution of Dirofilaria

and dirofilariosis in Europe: state of the art. In: Proceedings

of the 2003 Helminthological Colloquium, Vienna, Austria, pp.

6–11., http://www.vu-wien.ac.at/i116/oegtp/downloads_oegtp/

Abstractband.pdf.

Huby-Chilton, F., Beveridge, I., Gasser, R.B., Chilton, N.B., 2001.

Single-strand conformation polymorphism analysis of genetic

variation in Labiostrongylus longispicularis from kangaroos.

Electrophoresis 22, 1925–1929.

Jackson, R.F., 1975. A filarial parasite in greyhounds in Southern

Ireland. Vet. Rec. 97, 476–477.

Knott, J., 1935. The periodicity of the mf of Wuchereia bancrofti,

preliminary report of some injection experiments. Trans. Roy.

Soc. Trop. Med. Hyg. 29, 59–64.

Manuali, E., Eleni, C., Giovannini, P., Costarelli, S., Ciorba, A.,

2005. Unusual finding in a nipple discharge of a female dog:

dirofilariasis of the breast. Diagn. Cytopathol. 32, 108–

109.

Mar, P.H., Yang, I.C., Chang, G.N., Fei, A.C., 2002. Specific

polymerase chain reaction for differential diagnosis of Dirofi-

laria immitis and Dipetalonema reconditum using primers

derived from internal transcribed spacer region 2 (ITS2). Vet.

Parasitol. 106, 243–252.

Nassau County Department of Health, 2003. Nassau County Mos-

quito Surveillance and Control Report.

Newton, L.A., Chilton, N.B., Beveridge, I., Gasser, R.B., 1998.

Systematic relationships of some members of the genera Oeso-

phagostomum and Chabertia (Nematoda: Chabertiidae) based

on ribosomal DNA sequence data. Int. J. Parasitol. 28, 1781–

1789.

Otto, G.F., 1977. The significance of microfilaremia in the diagnosis

of heartworm infection. In: Proceedings of the American

Heartworm Society Heartworm Symposium. pp. 22–30.

Pacheco, G., Tulloch, G.S., 1970. Microfilariae of Dirofilaria striata

in a dog. J. Parasitol. 56, 248.

Pampiglione, S., Rivasi, F., Angeli, G., Boldorini, R., Incensati,

R.M., Pastormerlo, M., Pavesi, M., Ramponi, A., 2001. Diro-

filariasis due toDirofilaria repens in Italy, an emergent zoonosis:

report of 60 new cases. Histopathology 38, 344–354.

Pampiglione, S., Rivasi, F., Paolino, S., 1996. Human pulmonary

dirofilariasis. Histopathology 29, 69–72.

Romano, A., Sachs, R., Lengy, J., 1976. Human ocular dirofilariasis

in Israel. Isr. J. Med. Sci. 12, 208–214.

Sambrook, J., Russell, D.W. (Eds.), 2001. Molecular Cloning: A

Laboratory Manual, 3rd ed., vol. 1. Cold Spring Harbor Labora-

tory Press, Cold Spring Harbor.

Tarello, W., 2003. Dermatitis associated with Dirofilaria repens

microfilariae in three dogs in Saudi Arabia. J. Small Anim. Pract.

44, 132–134.

Vakalis, N., Spanakos, G., Patsoula, E., Vamvakopoulos, N.C., 1999.

Improved detection of Dirofilaria repens DNA by direct poly-

merase chain reaction. Parasitol. Int. 48, 145–150.

Vakalis, N.C., Himonas, C.A., 1997. Human and canine dirofilar-

iasis in Greece. Parassitologia 39, 389–391.

Vasilkova, D., Klisenbauer, D., Juhas, T., Moravec, F., Uhlikova, M.,

Hubner, J., Konakova, G., 1992. Isolation of Dirofilaria repens

in vitreoretinal findings. Cesk. Oftalmol. 48, 274–277.

Woods, W.G., Whithear, K.G., Richards, D.G., Anderson, G.R.,

Jorgensen, W.K., Gasser, R.B., 2000. Single-strand restriction

fragment length polymorphism analysis of the second internal

transcribed spacer (ribosomal DNA) for six species of Eimeria

from chickens in Australia. Int. J. Parasitol. 30, 1019–

1023.

Zahler, M., Glaser, B., Gothe, R., 1997. Imported parasites in dogs:

Dirofilaria repens and Dipetalonema reconditum. Tierarztl.

Prax. 25, 388–392.