Capillary Tube Feeding System for Studying Tick ... - Oxford Journals

2 downloads 0 Views 475KB Size Report
Dermacentor andersoni Stiles with differences in size and weight (de la Fuente et al. 2005) and to test the effect of antibodies against tick-protective antigens.
VECTOR/PATHOGEN/HOST INTERACTION, TRANSMISSION

Capillary Tube Feeding System for Studying Tick–Pathogen Interactions of Dermacentor variabilis (Acari: Ixodidae) and Anaplasma marginale (Rickettsiales: Anaplasmataceae) KATHERINE M. KOCAN,1 JOY YOSHIOKA,1 DANIEL E. SONENSHINE,2 JOSE´ DE LA FUENTE,1, 3 SHANE M. CERAUL,2 EDMOUR F. BLOUIN,1 AND CONSUELO ALMAZA´N1

J. Med. Entomol. 42(5): 864Ð874 (2005)

ABSTRACT A capillary tube feeding (CTF) system was adapted for studying the interaction between Dermacentor variabilis (Say) and the rickettsial cattle pathogen Anaplasma marginale Theiler. A. marginale undergoes a complex developmental cycle in ticks that begins in midguts and ends by transmission from salivary glands. In this CTF system, male D. variabilis were fed A. marginale-infected blood or cultured tick cells. Ticks that fed on highly rickettsemic calves developed midgut and salivary gland infections as detected by PCR, whereas ticks that were fed from capillary tubes on the same blood developed only midgut infections. An unexpected result of capillary tube feeding was that antibodies against the A. marginale adhesin, major surface protein 1a, enhanced midgut infections and caused cell culture-derived A. marginale to infect midguts. Another unexpected result was the infection of the midguts of the nonvector tick Amblyomma americanum (L.), after capillary tube feeding on infected bovine blood. The gut cell response of ticks to A. marginale, as determined from SDS-polyacrylamide gel electrophoresis protein proÞles, did not differ when ticks were fed infected or uninfected cells from capillary tubes. Selected protein bands, as identiÞed by tryptic digestion-mass spectrometry, contained mostly proteins of bovine origin, including bovine albumin, undigested ␣- and ␤-chain hemoglobin and hemoglobin fragments. Although infection of ticks by A. marginale CTF system was not the same as infection by feeding on cattle, the results obtained demonstrated the potential use of this system for identifying aspects of pathogenÐvector interactions that are not readily recognized in naturally feeding ticks. KEY WORDS Anaplasma marginale, MSP1a, Dermacentor variabilis, Amblyomma americanum, capillary tube feeding

TICKS ARE OBLIGATE PARASITES that feed exclusively on the blood of their vertebrate hosts. The tick midgut is the Þrst barrier that must be traversed by pathogens ingested with the bloodmeal. Infection must be established Þrst in midguts before the pathogen can subsequently move to other tissues, most notably the salivary glands, which is the site of transmission to the vertebrate host. During feeding, the midgut is exposed to a variety of nutritive elements (e.g., hemoglobin) as well as noninfective bacteria, such as Bacillus subtilis, from host skin, and various pathogenic rickettsia, bacteria, viruses, and protozoa with tick vectors (Sonenshine 1993). The relationship between ticks and pathogens seems to be well established because infections are limited and generally not injurious to the tick. The mechanism by which invasive microorgan1 Department of Veterinary Pathobiology, Center for Veterinary Health Sciences, Oklahoma State University, Stillwater, OK 740782007. 2 Department of Biological Sciences, Old Dominion University, Norfolk, VA 23529-0266. 3 Instituto de Investigacio ´ n en Recursos Cinege´ ticos, Ronda de Toledo s/n, 13005 Ciudad Real, Spain.

isms colonize tick tissues without provoking an effective immune defense is not well understood. A capillary tube feeding (CTF) system offers a method of exposing ticks to pathogens without the use of infected hosts and provides an artiÞcial system in which the composition of the tick meals could be varied for experimental purposes. A CTF system, used initially as a feeding system for soft ticks, was adapted to infect ixodid ticks with rickettsial organisms (Rau and Hannoun 1968; Rechav et al. 1999; Macaluso et al. 2001). In the case of D. variabilis, rickettsial infection was systemic, resulting in transmission by salivary gland and transovarial transmission mechanisms (Macaluso et al. 2001). A CTF system also was developed recently by Broadwater et al. (2002) as an alternate way of infecting nymphal Ixodes scapularis Say with Borrelia burgdorferi, the causative agent of Lyme disease. Capillary fed ticks transmitted B. burgdorferi to mice, although not as efÞciently as naturally infected ones. The CTF system was subsequently used to study expression of defensin-like peptides of D. variabilis in response to exposure to B. burgdorferi (Sonenshine et al. 2002), to evaluate the volume of ingested blood of

0022-2585/05/0864Ð0874$04.00/0 䉷 2005 Entomological Society of America

September 2005

KOCAN ET AL.: CAPILLARY FEEDING STUDIES OF TICKS AND A. marginale

Dermacentor andersoni Stiles with differences in size and weight (de la Fuente et al. 2005) and to test the effect of antibodies against tick-protective antigens that were identiÞed in I. scapularis by expression library immunization (Almaza´n et al. 2005). The focus of this research was to adapt the CTF system for infection of D. variabilis with the cattle pathogen Anaplasma marginale to provide an alternate system for studying tickÐpathogen interactions. The development of A. marginale in Dermacentor spp. involves a complex cycle that begins in the midgut, followed by infection of several tick tissues, including salivary glands, which are the site of transmission to cattle (Kocan et al. 1992). The process of infection of tick cells by A. marginale was shown to be initiated by adhesion of the rickettsiae to the host cell membrane (Blouin and Kocan 1998). We subsequently demonstrated that major surface protein (MSP) 1a was the A. marginale adhesion protein for tick cells (de la Fuente et al. 2001a). The extracellular N-terminal region of MSP1a consists of tandemly repeated peptides that contain adhesion moieties (de la Fuente et al. 2001c, 2003a, c), B-cell epitopes (Garcia-Garcia et al. 2004a), and a neutralization-sensitive epitope (Palmer et al. 1987). Antibodies to MSP1a reduced A. marginale infections in cultured tick cells and ticks (Blouin et al. 2003; de la Fuente et al. 2003b). MSP1a was found to be differentially expressed in A. marginale from bovine erythrocytes and from ticks cells (de la Fuente et al. 2001a, b; Garcia-Garcia et al. 2004b), being down regulated in A. marginale derived from tick cells, which may inßuence the ability of ticks to become infected with A. marginale from cultured tick cells. The research presented herein was designed to determine whether the CTF system would provide a model system for experimental study of the A. marginaleÐtick interface. Ticks were capillary tube fed A. marginale-infected bovine blood and cultured tick cells, and infection of tick midguts and salivary glands was tested by PCR. Infection of ticks by capillary tube feeding and by feeding on cattle was compared. We were most interested to determine whether the tick cell culture-derived A. marginale, in which MSP 1a was down-regulated, were infective for ticks. The effect of rMSP1a antibodies on infectivity of A. marginale from bovine erythrocytes or tick cell culture also was studied. Finally, we sought to compare the tick midgut protein response of ticks capillary fed compared with ticks fed on cattle. We also were interested to determine whether antimicrobial peptides were produced by ticks in response to exposure to A. marginale, as was demonstrated when other blood-feeding arthropods were exposed to other microbes (for review, see Nakajima et al. 2002). Materials and Methods Ticks. Male D. variabilis and Amblyomma americanum (L.) used for these studies were reared at the Department of Entomology Tick Rearing Facility, Oklahoma State University. A. americanum, a nonvector tick species for A. marginale, was included as an

865

additional control for the tick feeding studies. Larvae and nymphs were fed on rabbits and held in humidity chambers until molting to the adult stage. Male ticks selected for the study were held at 25⬚C, 90 Ð98% RH, and photoperiod of 14:10 (L:D) h (Kocan et al. 1992). A. marginale Isolates. Four isolates (Oklahoma, Stillwater 68, Virginia, and Okeechobee isolates) of A. marginale, derived originally from infected cattle, were used for these studies. The Oklahoma, Stillwater 68, and Virginia isolates were shown to be infective and transmissible by ticks, whereas the Okeechobee isolate was proven in previous studies to be nontransmissible by ticks and not infective for cultured tick cells (Blouin et al. 2000, 2002; de la Fuente et al. 2001b, 2003a). The Virginia isolate, infective for cattle and ticks, was cultivated in cultured tick cells for use in these CTF studies (Munderloh et al. 1996; Blouin et al. 1998). Propagation of A. marginale in Tick Cell Culture. The Virginia isolate of A. marginale was propagated in the tick cell line IDE8 (ATCC CRL 11973), derived from I. scapularis embryos, as described previously (Munderloh et al. 1996; Blouin et al. 1998, 2000, 2002). Brießy, tick cells were maintained at 31⬚C in L-15 B medium, pH 7.2, supplemented with 5% heat inactivated fetal bovine serum (Sigma, St. Louis, MO), 10% tryptose phosphate broth (Difco, Detroit, MI), and 0.1% lipoprotein concentrate (MP Biomedicals, Irvine, CA), and the culture medium was replaced weekly. Monolayers of IDE8 cells were inoculated with the Virginia isolate of A. marginale and monitored for rickettsial infection by examination of stained smears (HEMA 3 stain set, Biochemical Sciences, Inc., Swedesburg, NJ) and with phase contrast microscopy. Infected cell cultures, in which ⬎80% of the cells were infected, were harvested by centrifugation (Fig. 1A and B). Expression of rMSP1a, Purification, and Antibody Preparation. The gene msp1␣ of the Oklahoma isolate of A. marginale, encoding for MSP1a, was cloned and expressed in Escherichia coli as reported previously (de la Fuente et al. 2001b). Expression of the recombinant MSP1a (rMSP1a) was conÞrmed by SDS-polyacrylamide gel electrophoresis (PAGE) (Laemmli 1970). The rMSP1a was puriÞed by FLAG-afÞnity chromatography (Sigma) following the manufacturerÕs instructions and used to immunize cattle as described by Garcia-Garcia et al. (2004b). rMSP1a antibody titers were determined by use of an enzymelinked immunosorbent assay (ELISA) as described by Garcia-Garcia et al. (2004b). Polyclonal bovine IgG was puriÞed from sera of MSP1a-immunized cattle and preimmune bovine sera by protein G afÞnity chromatograph as described by Blouin et al. (2003). Experimental Cattle and Tick/Cattle Feeding. Nine mixed breed calves (8 Ð10 mo old), determined to be free of infection by an A. marginale-speciÞc competitive ELISA (Saliki et al. 1998), were used for as a source of infected or uninfected blood for tickmeals and for acquisition feeding of ticks. Cattle were housed in the Anaplasmosis Research Barn and cared for by the Oklahoma State University Laboratory An-

866

JOURNAL OF MEDICAL ENTOMOLOGY

Vol. 42, no. 5

Fig. 1. Light micrographs of cultured tick cells infected with A. marginale that were combined with uninfected bovine blood and capillary fed to ticks. Stain, HEMA 3 stain set. (A) Two colonies of A. marginale in an intact cultured cell (arrows). (B) A. marginale released from cultured tick cells (arrow). Bar, 10 ␮m.

imal Research Unit with the approval the Institutional Animal Care and Use Committee, Protocol No. VM 50802. Calves experimentally infected with A. marginale were monitored three times a week by examination of stained blood smears and determination of the packed cell volume. Thin smears were made on glass slides from cattle blood collected by venipuncture in EDTA-treated Monoject blood collection tubes (Sherwood Medical, St. Louis, MO) and stained with Protocol Hem 3 stain (Biochemical Sciences). The percentage of infected erythrocytes out of 500 was determined and reported as the percentage of infected erythrocytes (PPE). Once infection was detected in blood smears, the calves were monitored daily. Male ticks were allowed to feed for 4 d on infected cattle to acquire A. marginale infection (acquisition feeding), whereas companion ticks were capillary tube fed for 4 d. These acquisition-fed ticks were placed in orthopedic stockinettes glued to the shaved side of cattle as described by Kocan et al. (1992) at various rickettsemias and allowed to feed for 4 d. In the cattle and capillary tube feeding studies, additional groups of ticks were allowed to feed a second time for 7 d on a sheep (noncomponent host) to cause infection of A. marginale in tick salivary glands. The CTF System. The CTF system was developed according to the procedures of Broadwater et al. (2002). Before capillary tube feeding, ticks were prefed on sheep for 3 d to initiate tick feeding. The ticks were removed from the sheep and then immobilized on tape attached to a microscope slide that was placed in a petri dish as illustrated in Fig. 2A. Capillary tubes containing infected or uninfected cells were Þtted over the tick mouthparts and immobilized on double sticky tape on the petri dish rim. After placement of the capillary tubes, the ticks were placed in a humidity chamber and allowed to feed for 4 d. The

capillary tubes were changed daily and replaced with tubes containing fresh tickmeal. Fluorescent Microspheres and Tissue Smears for Confirmation of Capillary Tube Feeding. Studies were done on ticks capillary tube fed uninfected blood with or without ßuorescent microspheres to determine their use as a tracer for feeding. After conÞrmation of the use of the microspheres as a tracer for tick feeding, 20 ␮l of 1-␮m ßuorescent microspheres (Molecular Probes, Eugene, OR) per milliliter of blood was added to the tickmeal (Fig. 2B). In each CTF experiment, Þve ticks were selected at the completion of day 4 of feeding and used to prepare tissue smears for detection of ßuorescent microspheres and conÞrmation of uptake of the tick meal. Tissue smears included fecal, midgut and hemolymph smears. The tissues were smeared in phosphate-buffered saline (PBS), Þxed in methanol, and then examined by epißuorescence microscopy for the presence of the ßuorescent microspheres. Preparation of Tickmeals for Capillary Tube Feeding. Tickmeals for capillary tube feeding included A. marginale infected and uninfected bovine blood and A. marginale infected and uninfected cultured tick cells. For capillary tube feeding of ticks with infected or uninfected blood, blood was collected in EDTAtreated tubes from infected cattle on each of the 4 d of tick feeding, and PPE was determined. Fluorescent microspheres (100 ␮l/ml) were added to the tick meal, and the capillary tubes were Þlled with blood and Þtted to the tick mouthparts in the CTF system. Three isolates of A. marginale were tested for their infectivity for D. variabilis in capillary tube feeding and cattle feeding trials as listed in Table 1, which included the two tick transmissible isolates (Oklahoma and Stillwater 68 isolates) and one nontransmissible isolate (Okeechobee isolate). The nonvector tick species for A. marginale, A. americanum, also was

September 2005

KOCAN ET AL.: CAPILLARY FEEDING STUDIES OF TICKS AND A. marginale

867

Fig. 2. Capillary tube feeding of male tick D. variabilis. Ten adult ticks, previously fed on a bovine or sheep, were detached and conÞned between two glass slides in a petri dish. Glass microcapillaries Þlled with the feeding medium (including ßuorescent microspheres) were placed over the mouthparts of each tick and secured to a layer of double sticky tape around the rim of the petri dish. (A) Photograph illustrating a group of 10 ticks in the CTF system feeding on infected bovine blood. The tick meal level before feeding is marked by an arrow). (B) Fluorescent microspheres in a midgut smear from a tick that had capillary fed for 4 d. Bar, 10 ␮m.

tested in the CTF system by using the Stillwater 68 isolate of A. marginale (Table 1, group 9). To compare capillary tube feeding of ticks on low versus high PPE-infected blood, capillary tube and calf/tick feedings were done at low PPEs, ranging from 0.4 to 3.4%, and then by using the same cattle, at high PPEs, ranging from 25.5 to 61.8%. These studies were done in two trials by using cattle PA490 and PA492 that were infected with the Stillwater 68 isolate (Table 1, groups 2Ð5). Uninfected bovine blood was collected from cattle that were serologically negative as determined by the A. marginale MSP5 cELISA and in which inclusion bodies were not detected in stained

blood smears. Experimental groups for these tick/ cattle and capillary tube feeding experiments were 1) ticks allowed to feed for 4 d on infected cattle; 2) ticks capillary tube fed infected bovine blood for 4 d (blood collected daily from the cattle used for feeding group 1 ticks); 3) ticks capillary tube fed uninfected bovine blood; and 4) unfed, uninfected ticks. For capillary tube feeding of ticks with A. marginale infected cultured IDE8 cells (Fig. 1A and B), 100 ␮l of A. marginale-infected tick cells harvested from one T-25 tissue culture ßask (2 ⫻ 107 cells/5 ml) was added to 900 ␮l of uninfected bovine blood and 20 ␮l of ßuorescent microspheres. Experimental groups for

Table 1. Infection of midguts and/or salivary glands from D. variabilis and A. americanum males after feeding on infected cattle or CTF on A. marginale infected bovine blood as determined by an A. marginale msp4 PCR assay

Trial no.

D. variabilis males 1 2 3 4 5 6 7 8 A. americanum malesb 9

A. marginale PPE during feeding on an infected calf or CTF

PCR tick guts

PCR tick salivary glands

Ticks fed on Calf PA 484 Oklahoma isolate Tick CTF blood from PA 484 Ticks fed on Calf PA490 Stillwater 68 isolate Ticks CTF blood from PA 490 Ticks fed on calf PA 490 Stillwater 68 isolate Ticks CTF blood from PA490 Ticks fed on Calf PA492 Stillwater 68 isolate Ticks CTF blood from PA 492 Ticks fed on Calf PA492 Stillwater 68 isolate Ticks CTF blood from PA 492 Ticks fed on Calf 499 Oklahoma isolate Ticks CTF blood from PA499 Ticks fed on Calf PA498 Stillwater 68 isolate Ticks CTF blood from PA498 Ticks fed on calf PA489 Okeechobee isolatea Ticks CTF blood from PA489

1.7Ð21.6 1.7Ð21.6 0.7Ð3.4 0.7Ð3.4 25.5Ð47.2 25.5Ð47.2 0.4Ð3.4 0.4Ð3.4 28.6Ð61.8 28.6Ð61.8 23.6Ð71.5 23.6Ð71.5 33.1Ð63.9 33.1Ð63.9 26.4Ð51.9 26.4Ð51.9

⫹ ⫹ ⫹ ⫺ ⫹ ⫹ ⫹ ⫺ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫺ ⫺

NA NA ⫺ ⫺ ⫹ ⫺ ⫺ ⫺ ⫹ ⫺ ⫹ ⫺ NA NA ⫺ ⫺

Ticks fed on Calf PA 498 Stillwater 68 isolate Ticks CTF blood from PA 498

33.1Ð63.9 33.1Ð63.9

⫺ ⫹

NA NA

A. marginale isolate

NA, not available. Okeechobee isolate was shown not to be infective for D. variabilis ticks (de la Fuente et al. 2001a). A. americanum is not a tick vector of A. marginale (for review, see Kocan et al. 2004).

a

b

868

JOURNAL OF MEDICAL ENTOMOLOGY

Vol. 42, no. 5

Table 2. Infection of D. variabilis midguts in ticks fee by CTF on a mixture of A. marginale-infected cultured tick cells and bovine whole blood with or without bovine rMSP1a IgG

Trial no. Ticks CTF-infected blooda and rMSP1a bovine IgG 1

2

Ticks CTF infected cultured tick cellsb and rMSP1a bovine IgG 3

A. marginale isolate/cell source/rMSP1a antibody

Ticks fed on Calf PA484Stillwater 68 isolate Tick CTF blood from PA484 and rMSP1a antibody undiluted (100 ␮l/ml) Tick CTF blood from PA484 and rMSP1a antibody 1:10 (10 ␮l/ml) Tick CTF blood from PA484 and rMSP1a antibody 1:100 (1 ␮l/ml) Ticks fed on calf PA499 Stillwater 68 isolate Ticks CTF blood from PA499 and rMSP1a antibody undiluted (100 ␮l/ml) Ticks CTF blood from PA499 and rMSP1a antibody 1:5 (20 ␮l/ml) Ticks CTF blood from PA499 and rMSP1a antibody 1:10 (10 ␮l/ml)

Ticks fed cultured tick cells infected with A. marginale (Virginia isolate) Ticks fed cultured tick cells infected with A. marginale (Virginia isolate) and rMSP1a antibody undiluted (100 ␮l/ml)

A. marginale rickettsemias PCR during calf feeding or tick guts CTF (%)

17.5Ð52.4 17.5Ð52.4 17.5Ð52.4 17.5Ð52.4 23.6Ð71.5 23.6Ð71.5 23.6Ð71.5 23.6Ð71.5

⫹ ⫹ ⫹ ⫺ ⫹ ⫹ ⫹ ⫹

⬎80 ⬎80

⫺ ⫹

Infection was determined by use of an A. marginale msp4 PCR assay. a Percentage of parasitized erythrocytes. b Percentage of infected cultured tick cells.

CTF studies with A. marginale-infected tick cell cultures included 1) infected tick cell cultures in uninfected blood, 2) uninfected cultured tick cells in uninfected bovine blood, and 3) uninfected bovine blood. Three replicates of the experiment were done. For CTF studies to test the effect of rMSP1a antibodies, blood from infected cattle PA484 and PA499 was mixed with bovine rMSP1a IgG (2.7 mg IgG/ml) at various dilutions (undiluted, 1:5, 1:10, and 1:100 as listed in Table 2, trials 1 and 2). Infected tick cell cultures containing undiluted bovine rMSP1a IgG also were capillary tube to fed male D. variabilis (Table 2, trial 3). The controls for these rMSP1a experiments included tick meals of 1) uninfected cultured cells, 2) uninfected bovine blood, and 3) infected tick cells with bovine preimmune sera puriÞed IgG (2.5 mg/ml). A. marginale msp4 Polymerase Chain Reaction (PCR) for Detection of A. marginale-Infected Tick Guts and Salivary Glands. A. marginale DNA was extracted from the guts or salivary glands dissected from 10 capillary tube or cattle fed ticks per experimental group and pooled in RNALater (Ambion, Austin, TX). Tick DNA was extracted and the msp4 gene, shown previously to be a stable marker of A. marginale, was ampliÞed as described by de la Fuente et al. (2003b, 2005). Tissue Collections and Protein Assays. After the 4-d feeding period on cattle or by CTF, ticks were washed; cut in half, separating the left and right sides; and the midguts were dissected in sterile PBS and placed in cold (4⬚C) protein collection buffer containing 100 mM PBS, 0.1Ð 0.2 mM phenylmethylsulfonyl ßuoride, and 1: 200 dilution of protease inhibitor cocktail (Sigma). The midgut samples were homogenized, and

the protein concentration was determined using the Bradford protein assay and immunoglobulin G as described by the manufacturer (Bio-Rad, Hercules, CA). SDS-PAGE. SDS-polyacrylamide gel electrophoresis was done using Tris-Bis 4 Ð12% gradient NuPage minigels, 10 cm by 10 cm by 1 mm thick (Invitrogen, Carlsbad, CA). Midgut samples collected as described above were adjusted to similar protein content (30 ␮g), and gel electrophoresis was done under reducing conditions in accordance with the manufacturerÕs recommendations. Gels were silver-stained (Silver Express, Invitrogen), photographed with a Kodak DC120 digital camera (Eastman Kodak, Rochester, NY), and relative molecular weights (rMW) were assigned using the Kodak ID software. Also included for comparison were protein samples made from lysed cultured IDE8 cells, both infected and uninfected. Identification of Selected Tick Midgut Protein Bands from SDS-PAGE Gels. Midgut protein samples from cattle or capillary tube fed ticks were separated into two fractions by using 50 molecular weight cut off Microcon Þlters (Millipore, Billerica, MA). After concentration (Lyphlock lyophilizer, LabConco, Kansas City, MO), each fraction was fractionated further using a Waters high-performance liquid chromatography (HPLC) system (Waters Corporation, Milford, MA) and a Shimadzu model SPD-M10A photodiode array detector (Shimadzu ScientiÞc Instruments, Columbia, MD). The column was a 250 mm by 4.6 mm i.d. Vydac reversed phase C4 protein column (Nest Group, Southborough, MA) containing 5-␮m silica particles (300 Å). The solvents were HPLC grade 0.1% trißuoroacetic acid (TFA) and acetonitrile (J.T. Baker, Phillipsburg, NJ). Run conditions were as follows: linear gradient from 85% TFA; 15%

September 2005

KOCAN ET AL.: CAPILLARY FEEDING STUDIES OF TICKS AND A. marginale

acetonitrile to 15% TFA; 85% acetonitrile over 60 min. Peaks separated by HPLC were collected using a Frac-100 fraction collector (Amsersham Biosciences Inc., Piscataway, NJ), concentrated, and monitored for protein content by gel electrophoresis as described above. Fractions puriÞed as described above were submitted to either the Oklahoma State University Biochemistry Core Facility or Biomedical Research Facility, W. M. Keck Biomedical Mass Spectrometry Laboratory located at the University of Virginia Health System for protein identiÞcation by tryptic digestion and sequencing by mass spectrometry. Peaks were identiÞed by matching with known sequences in the National Center for Biotechnology Information protein database. Matrix-assisted laser desorption-time of ßight mass spectrometry was used to determine the absolute molecular weights of the peptides or peptide fragments.

Results Confirmation of Tick Feeding from Capillary Tubes. In preliminary studies, the shape and size of the ßuorescent microspheres were found to be distinct and easily recognized in tissues of ticks fed uninfected blood. Fluorescent artifacts similar in size and shape to the microspheres were not observed in ticks capillary tube fed uninfected blood without the microspheres. In the CTF studies, ßuorescent microspheres were detected in all tick tissue smears (hemolymph, midgut, and fecal) from capillary fed ticks, thus conÞrming that ticks had fed from the capillary tubes (Fig. 2B; data not shown). PCR of Midguts from Ticks Fed A. marginale-Infected Bovine Blood. The PCR results of midguts and salivary glands dissected from ticks exposed to A. marginale-infected blood by capillary tube feeding or feeding on infected cattle are listed in Table 1. Midguts of ticks that were allowed to feed on cattle infected with the Oklahoma or Stillwater 68 isolates were all PCR positive for A. marginale, whereas those from ticks that fed on a calf infected with the Okeechobee isolate were PCR negative for A. marginale. Only midguts from ticks capillary tube fed on highly rickettsemic blood of the Oklahoma or Stillwater 68 isolates proved to be PCR positive for A. marginale, whereas midguts from ticks capillary tube fed on blood with low rickettsemias were PCR negative for A. marginale. When tick salivary glands were tested by PCR after ticks were allowed to feed a second time on sheep to cause development of A. marginale in tick salivary glands, the only salivary glands samples that were PCR positive for A. marginale were those dissected from ticks that acquired infection by feeding on cattle with high A. marginale rickettsemias. A. marginale salivary gland infections were not detected by PCR in groups of ticks that were capillary tube fed on infected blood and were then allowed to feed a second time on sheep (trials 3Ð6). When A. americanum were allowed to feed on a cow infected with the Stillwater 68 isolate of A. marginale, the midguts dissected from the ticks were

869

PCR negative for A. marginale. However, when blood from the same calf, PA498, was capillary tube fed to A. americanum, the midguts dissected from those ticks were A. marginale PCR positive (Table 1, trial 9). PCR of Midguts and Salivary Glands from Ticks Capillary Tube Fed A. marginale-Infected Cell Cultures. In three trials, when D. variabilis were capillary tube fed cell cultures infected with the Virginia isolate of A. marginale mixed with uninfected bovine blood, A. marginale infection in midguts was not detected by PCR. PCR Results of Ticks Fed from Capillary Tubes Containing rMSP1a IgG with Infected Blood or Cell Cultures. The PCR results of midguts from ticks that were fed A. marginale-infected bovine blood or infected tick cell cultures from capillary tubes are listed in Table 2. In two trials, addition of undiluted bovine rMSP1a IgG to A. marginale-infected blood resulted in greater A. marginale infection of tick midguts as determined by PCR. In trial 3, ticks fed infected tick cell cultures from capillary tubes (which proved in previous experiments to be noninfective for ticks) and undiluted rMSP1a IgG, resulted in ticks with midguts that were PCR positive for A. marginale. Tick Midgut Protein Profiles after Capillary Tube Feeding and Identification of Selected Protein Bands. Protein proÞles from male D. variabilis ticks that were allowed to feed on A. marginale-infected cattle with high or low rickettsemias or fed from capillary tubes on the same infected bovine blood are shown in Fig. 3. Protein bands from the calf and capillary tube fed tick groups were similar as determined by visual inspection of the stained gels and digital camera densitometry measurements, except that in the tick midgut proÞles larger bands from capillary fed ticks were observed that comigrated with putative lysozyme (13.9 kDa). These bands were subsequently shown by tryptic digestion-mass Þngerprinting to be the ␣- and ␤-chain moieties of hemoglobin (data not shown). A noteworthy Þnding was that the ␣- and ␤-chain hemoglobin bands in the ticks exposed to high rickettsemia blood were approximately twice as large as comparable bands from the ticks fed low rickettsemia blood in capillary tubes (Fig. 3A and B). Bands observed at ⬇5.5 kDa, Þrst thought to be defensin, proved by mass spectrometry-mass Þngerprinting to be fragments of ␣- and ␤-chain hemoglobin (Hb fragments). Protein Profiles and Mass Spectrometry of A. americanum Fed from Capillary Tubes. Protein proÞles of midguts trial 9 (Table 1) dissected from D. variabilis or A. americanum males after calf feeding or capillary tube feeding were comparable, except that selected bands, subsequently shown by mass spectrometry to be bovine albumin and ␣- and ␤-chain hemoglobin, were larger in the A. americanum ticks that were fed from capillary tubes (Fig. 4, lanes 6 and 7). Proteins found by tryptic digestion-mass Þngerprinting included 1) bovine transferin and bovine albumin (Fig. 4, top marked band); 2) tick cysteine and serine proteases as well as bovine ␣- and ␤-chain hemoglobin, transthyretin, albumin, IgG, cytokeratin, and Þbrino-

870

JOURNAL OF MEDICAL ENTOMOLOGY

Vol. 42, no. 5

Fig. 3. SDS-PAGE gels comparing the tick midgut protein response to low levels versus high levels of A. marginaleinfection. (A) Ticks exposed to A. marginale infected blood with low rickettsemias. Lanes 1 and 2, ticks acquisition-fed on bovine PA 490 and PA492, respectively; lanes 3 and 5, ticks capillary fed A. marginale-infected blood from PA 490 and PA492, respectively; lane 4, molecular weight standards; and lane 6, ticks capillary fed on an uninfected bovine blood; and lane 7, uninfected, unfed ticks. Bands identiÞed as ␣- and ␤-chain hemoglobin, undigested (large arrowhead). Bands identiÞed as fragments of tick digestion of hemoglobin (small arrowhead). (B) Ticks exposed to A. marginale-infected blood with high rickettsemias. Lanes 1 and 2, ticks fed on infected calves, PA490 and PA 492, respectively; lane 3, molecular weight markers; lanes 4 and 5, ticks capillary tube fed on A. marginale-infected bovine blood from PA 490 and PA 492, respectively; and lane 6, ticks capillary tube fed infected blood without the addition of ßuorescent microspheres. Bands identiÞed as ␣- and ␤-chain hemoglobin, undigested (large arrowhead). Bands identiÞed as fragments of tick digestion of hemoglobin (small arrowhead).

gen (Fig. 4, bottom marked band); and digestive fragments of bovine ␣- and ␤-chain hemoglobin. Discussion The CTF system proved to be effective for infecting D. variabilis males with A. marginale, although many

experiments were required before a format was obtained that made it possible to achieve this result. After capillary tube feeding, attachment and feeding on sheep was required for development of A. marginale in tick salivary glands. Blood was found to be a necessary component in the tickmeal. PBS; bovine serum, whether or not heat inactivated; tick cells in culture

September 2005

KOCAN ET AL.: CAPILLARY FEEDING STUDIES OF TICKS AND A. marginale

871

Fig. 4. SDS-PAGE gels comparing the tick midgut protein response of D. variabilis and A. americanum to bovine blood infected with the Stillwater 68 isolate of A. marginale by feeding on a calf PA498 or from blood in capillary tubes. Lane 1, D. variabilis fed on the infected calf; lane 2, D. variabilis capillary tube fed on blood from the infected calf; lane 3, molecular weight markers; lane 4, D. variabilis capillary tube fed on uninfected blood; lane 5, A. americanum fed on the infected calf; lane 6, A. americanum capillary tube fed on blood from the infected calf; and lane 7, A. americanum capillary tube fed uninfected bovine blood. White asterisks in Lane six indicate bands excised and subsequently identiÞed by tryptic digestionmass Þngerprinting.

medium; or the culture medium alone did not facilitate tick survival; ticks died either during the 4-d capillary tube feeding period or during infestation or feeding on sheep. Feeding success and survival was greatly improved when the ticks were allowed to prefeed on sheep before being fed from capillary tubes. In the CTF system, tick feeding was conÞrmed by visual decrease of the tick meal level in the capillary tubes and by conÞrmation of the ßuorescent microspheres in tick hemolymph, fecal, and midgut smears done after the fourth day of capillary tube feeding. Use of the CTF system for infection of D. variabilis males with A. marginale led to several surprising differences in the infectivity of A. marginale, discussed as follows, that were not observed with normal feeding. Exposure of ticks to A. marginale-infected blood by capillary tube feeding resulted in midgut infections as detected by the msp4 PCR assay, but only when high rickettsemic blood was fed to the ticks. The ticks that

calf fed on the Oklahoma or Stillwater 68 isolates of A. marginale were all PCR-positive, regardless of the rickettsemia at the time of feeding. These results seem to be valid because ticks exposed to a nontransmissible control isolate, the Okeechobee isolate, either by calf or capillary tube feeding, had PCR-negative midguts. Because low rickettsemic blood was sufÞcient to initiate infection in ticks fed on cattle but not by capillary tube feeding, it is likely that some component(s) present in either the tick saliva or the host environment are missing in the CTF system. Infections of ticks exposed to low rickettsemic blood may have been prevented by immune defenses of the ticks, either by feeding on cattle or from capillary tubes. Alternately, tick midgut infection levels may have been below the detection level for PCR. None of the ticks fed from capillary tubes developed salivary gland infections, whereas the ticks that acquired infection on calves with high rickettsemias had PCR positive salivary

872

JOURNAL OF MEDICAL ENTOMOLOGY

glands. Although tick midguts became infected after ticks were fed from capillary tubes, these results collectively demonstrated that this artiÞcial feeding system differed and did not duplicate the natural process of infection when ticks fed on A. marginale-infected cattle; midgut infections acquired by capillary tube feeding apparently did not initiate the development cycle that ends in salivary gland infections. Because the capillary tube fed ticks received the same blood/ rickettsemias daily as the calf fed ticks, these differences may be due to alteration of the microbe, tick feeding physiology, and/or midgut expressed proteins. Another unexpected Þnding of this research was the ability of capillary feeding of ticks to induce A. marginale infection in midguts of A. americanum males. This species is not a vector for A. marginale, a fact conÞrmed by our Þnding that A. americanum males were refractory to infection when fed on an infected calf. Although changes in the feeding mechanism that may result from the CTF system are not known, feeding ticks from capillary tubes seems to have changed the tickÐpathogen interface. One of our main objectives for developing the CTF system was to determine whether A. marginale derived from cultured tick cells was infective for ticks. Transmission of pathogens from infected to uninfected ticks while feeding at the same site, called nonsystemic transmission by Randolph et al. (1996), was Þrst described for Thogoto virus (Jones et al. 1987) and subsequently demonstrated for several other viruses (as reviewed by Randolph et al. 1996). Nonsystemic transmission of the spirochete, Borrelia burgdorferi, was described by Gern and Rais (1996) in which larval Ixodes ricinus L. acquired infection with B. burgdorferi when infected nymphs fed at the same site. Piesman et al. (1998) also demonstrated nonsystemic transmission of B. burgodroferi from Þeld-collected female ticks to uninfected laboratory reared ones. We were not surprised when the tick cell culture-derived organisms were not infective for ticks by feeding from capillary tubes because we demonstrated in previous studies that expression of the A. marginale tick cell adhesion protein, MSP1a, is downregulated in tick cells (de la Fuente et al. 2001a, b; Garcia-Garcia et al. 2004b). The result of this study also was supported by a previous study in which we demonstrated that A. marginale was not transmitted from infected to uninfected ticks that co-fed together at the same site (Kocan and de la Fuente 2003). Thus, although some rickettsiae grown in mammalian cells (e.g., R. montanensis grown in Vero cells) can be transmitted successfully to ticks by feeding through capillary tubes (Macaluso et al. 2001), this apparently does not occur with A. marginale from cultured tick cells. However, the feeding of R. montanensis on cultured tick cells from capillary tubes has not been reported. An unexpected result of this research was the infection enhancing effect rather than the expected blocking effect that occurred when rMSP1a IgG was added to the tick meals. Undiluted IgG combined with

Vol. 42, no. 5

highly rickettsemic blood caused a notably greater PCR reaction in midgut samples. Likewise, inclusion of rMSP1a IgG in the tick meals with A. marginale infected tick cell cultures changed these cultures from noninfective to infective as determined by PCR. We can only speculate that addition of the rMSP1a IgG effected phagocytosis of the organisms by the epithelial cells of the midgut. However, whether they remained viable after their internalization is unknown. In contrast, we demonstrated previously that inclusion of rMSP1a IgG reduced A. marginale infections in cultured tick cells and in ticks that fed on immunized cattle (Blouin et al. 2003, de la Fuente et al. 2003b). Recent advances concerning the immune response of ticks to pathogens have shown that the tick immune response involves antimicrobial peptides, i.e., defensin, as well as lysozyme, ␣ and ␤-chain hemogloblin, and hemoglobin fragments that result from tick digestion of the bloodmeal (Johns et al. 2000, 2001a, b; Ceraul et al. 2003). Evidence of defensin expression was found in the midguts of soft ticks, Ornithodoros moubata Murray sensu latu, in response to blood feeding (Nakajima et al. 2001, 2002), but not in the midguts of D. variabilis (D.E.S. et al., unpublished data). However, in view of the concomitant occurrence of hemoglobin fragments with known antimicrobial activity (Fogaca et al. 1999, Nakajima et al. 2003), its role in innate immunity is unclear. Tick exposure to A. marginale-infected blood did not result in the appearance of unique protein bands when the midgut protein proÞles were examined by gel electrophoresis. No evidence of defensin or lysozyme was found. Bands that comigrated with lysozyme were identiÞed by mass spectrometry-mass Þngerprinting as bovine albumin and the undigested ␣- and ␤-chain moieties of hemoglobin. These bands were larger in the protein proÞles of capillary fed ticks than in those ticks that were allowed to feed naturally on cattle. Interestingly, the intensity of these bands was much greater in ticks capillary tube fed high compared with low rickettsemic blood (Fig. 3A and B). In addition, low molecular weight bands, Þrst thought to be a defensin, proved to be fragments of ␣and ␤-chain hemoglobin resulting from tick digestion of hemoglobin, similar to that reported in other tick species (Fogaca et al. 1999, Froidevaux et al. 2001, Nakajima et al. 2003). The intensity of these bands in capillary fed versus calf fed ticks was similar for ticks exposed to infected or uninfected blood. The appearance of larger albumin/hemoglobin protein bands in the tick gut protein proÞles from capillary fed ticks again suggests different mechanisms between capillary tick feeding (e.g., more efÞcient blood protein uptake) versus tick feeding on calves. The lack of a selective immune response in the tick midgut in response to A. marginale challenge was surprising but not necessarily unusual. In some bloodfeeding insects, e.g., the biting ßy Stomoxys calcitrans (L.), midgut defensins are constitutively produced but are up-regulated in response to lipopolysaccharide, part of the surface coat of gram-negative bacteria (Lehane et al. 1997). In mosquitoes, transcription of

September 2005

KOCAN ET AL.: CAPILLARY FEEDING STUDIES OF TICKS AND A. marginale

defensins occurs in the midguts not only after an infectious bloodmeal but also after noninfectious blood feeding (Lowenberger et al. 1999). Similarly, lysozyme was up-regulated strongly in the anterior midgut of the reduviid bug Triatoma infestans Klug immediately after blood feeding (Kollien et al. 2003). In the soft tick, O. moubata, blood feeding as well as bacterial challenge induced up-regulation of defensin (Nakajima et al. 2002), whereas blood feeding alone led to strong up-regulation of lysozyme at the transcriptional level (Grunclova et al. 2003). In summary, we developed a capillary tube feeding system for infection of ticks with A. marginale, and tick feeding was conÞrmed by detection of ßuorescent microspheres in tick tissue smears. Blood was found to be a necessary component of the tick meal required for tick survival. Although the resulting A. marginale midgut infections did not lead to infection of salivary glands, this system does provide a means for studying tick/pathogen interactions at the tick/midgut interface. Because the blood tickmeals were derived from the same cattle used for natural tick feeding, it is likely that the capillary tube feeding resulted in differences in tick gene expression, immunity, and midgut function. Differential gene expression when microbes are transferred from one cell type to another has been shown to occur in A. marginale (Garcia-Garcia et al. 2004b). Whether similar changes occur in ticks fed from capillary tubes is unknown. Although unique proteins were not produced in response to feeding on infected blood, nonspeciÞc tick defense mechanisms may have prevented infections in ticks that were capillary tube fed bovine blood with low A. marginale rickettsemias. The CTF system therefore provides an alternate system for experimentally studying the tick midgutÐA. marginale interface. Acknowledgments Steve Hartson (Core Sequencing Facility, Department of Biochemistry and Molecular Biology, Noble Research Center, Oklahoma State University, Stillwater, OK) is acknowledged for DNA sequencing. This research was supported by the project No.1669 of the Oklahoma Agricultural Experiment Station, the Endowed Chair for Food Animal Research (K.M. Kocan, College of Veterinary Medicine, Oklahoma State University) and National Institutes of Health Centers for Biomedical Research Excellence through a subcontract to J. de la Fuente from the Oklahoma Medical Research Foundation, and the Oklahoma Center for the Advancement of Science and Technology, Applied Research Grants AR00(1)001 and ARO21Ð37. OSU Microarray Core and Bioinformatics Resource Facilities are supported by grants from National Science Foundation (EOS-0132534) and National Institutes of Health (1P20RR16478-02 and 5P20RR15564-03). C.A. is supported by grants-in-aid from the Consejo Nacional de Ciencia y Tecnologia and Promep (University of Tamaulipas), Mexico, and a grant from PÞzer Animal Health, Inc., Kalamazoo, MI.

References Cited Almaza´ n, C., U. Blas-Machado, K. M. Kocan, J. H. Yoshioka, E. F. Blouin, A. J. Mangold, and J. de la Fuente. 2005.

873

Characterization of three Ixodes scapularis cDNAs protective against tick infestations. Vaccine. 23: 4403Ð 4416. Blouin, E. F., and K. M. Kocan. 1998. Morphology and development of Anaplasma marginale (Rickettsiales: Anaplasmataceae) in cultured Ixodes scapularis (Acari: Ixodidae) cells. J. Med. Entomol. 35: 788 Ð797. Blouin, E. F., A. F. Barbet, K. M. Kocan, and J. T. Saliki. 2000. Establishment and characterization of an Oklahoma isolate of Anaplasma marginale in cultured Ixodes scapularis cells. Vet. Parasitol. 87: 301Ð313. Blouin, E. F., J. de la Fuente, J. C. Garcia-Garcia, J. R. Sauer, J. T. Saliki, and K. M. Kocan. 2002. Applications of a cell culture system for studying the interaction of Anaplasma marginale with tick cells. Anim. Health Res. Rev. 3: 57Ð 68. Blouin, E. F., J. T. Saliki, J. de la Fuente, J. C. Garcia-Garcia, and K. M. Kocan. 2003. Antibodies to Anaplasma marginale major surface proteins 1a and 1b inhibit infectivity for cultured tick cells. Vet. Parasitol. 111: 247Ð260. Broadwater, A. H., D. E. Sonenshine, W. L. Hynes, S. Ceraul, and A. M. de Silva. 2002. Glass capillary tube feeding: a method for infecting nymphal Ixodes scapularis (Acari: Ixodidae) with the Lyme disease spirochete Borrelia burgdorferi. J. Med. Entomol. 39: 285Ð292. Ceraul, S. M., D. E. Sonenshine, R. E. Ratzlaff, and W. L. Hynes. 2003. An arthropod defensin expressed by the hemocytes of the American dog tick, Dermacentor variabilis (Acari: Ixodidae). Insect Biochem. Mol. Biol. 33: 1099 Ð1103. de la Fuente, J., J. C. Garcia-Garcia, E. F. Blouin, B. R. McEwen, D. Clawson, and K .M. Kocan. 2001a. Major surface protein 1a effects tick infection and transmission of the ehrlichial pathogen Anaplasma marginale. Int. J. Parasitol. 31: 1705Ð1714. de la Fuente, J., J. C. Garcia-Garcia, E. F. Blouin, and K. M. Kocan. 2001b. Differential adhesion of major surface proteins 1a and 1b of the ehrlichial cattle pathogen Anaplasma marginale to bovine erythrocytes and tick cells. Int. J. Parasitol. 31: 145Ð153. de la Fuente, J., J. C. Garcia-Garcia, E. F. Blouin, S. D. Rodriguez, M. A. Garcia, and K. M. Kocan. 2001c. Evolution and function of tandem repeats in the major surface protein 1a of the ehrlichial pathogen Anaplasma marginale. Anim. Health Res. Rev. 2: 163Ð173. de la Fuente, J., J. C. Garcia-Garcia, E. F. Blouin, and K. M. Kocan. 2003a. Characterization of the functional domain of the major surface protein 1a involved in adhesion, infection and transmission of Anaplasma marginale. Vet. Microbiol. 91: 265Ð283. de la Fuente, J., K. M. Kocan, J. C. Garcia-Garcia, E. F. Blouin, T. Halbur, and V. Onet. 2003b. Antibodies to Anaplasma marginale major surface protein 1a reduce infectivity for ticks. J. Appl. Res. Vet. Med. 1: 285Ð292. de la Fuente, J., R. A. Van Den Bussche, T. Prado, and K. M. Kocan. 2003c. Anaplasma marginale major surface protein 1␣ genotypes evolved under positive selection pressure but are not markers for geographic isolates. J. Clin. Microbiol. 41: 1609 Ð1616. de la Fuente, J., C. Almaza´ n, R. A. Van Den Busche, J. Bowman, J. H. Yoshioka, and K. M. Kocan. 2005. Characterization of genetic diversity in two populations of Dermacentor andersoni (Acari: Ixodidae). Exp. Parasitol. 109: 16 Ð26. Fogaca, A. C., P. I. da Silva, Jr., M. T. Miranda, A. G. Bianchi, A. Miranda, P. E. Ribolla, and S. Daffre. 1999. Antimicrobial activity of a bovine hemoglobin fragment in the tick Boophilus microplus. J. Biol. Chem. 274: 25330 Ð25334. Froidevaux, R., F. Krier, N. Nedjar-Arroume, D. VercaigneMarko, E. Kosciarz, C. Ruckebusch, P. Dhulster, and

874

JOURNAL OF MEDICAL ENTOMOLOGY

D. Guillochon. 2001. Antibacterial activity of a pepsinderived bovine hemoglobin fragment. FEBS Lett. 491: 159 Ð163. Garcia-Garcia, J. C., J. de la Fuente, K. M. Kocan, E. F. Blouin, T. Halbur, V. Onet, and J. T. Saliki. 2004a. Mapping of B-cell epitopes in the N-terminal repeated peptides of the Anaplasma marginale major surface protein 1a and characterization of the humoral immune response of cattle immunized with recombinant and whole organism antigens. Vet. Immunol. Immunopathol. 98: 137Ð151. Garcia-Garcia, J. C., J. de la Fuente, E. F. Blouin, T. Halbur, V. Onet, J. T. Saliki, and K. M. Kocan. 2004b. Differential expression of the msp1␣ gene of Anaplasma marginale occurs in bovine erythrocytes and tick cells. Vet. Microbiol. 98: 261Ð272. Gern, L., and E. Rais. 1996. EfÞcient transmission of Borrelia burgdorferi between cofeeding Ixodes ricinus ticks (Acari: Ixodidae). J. Med. Entomol. 33: 189 Ð192. Grunclova, L., H. Fouguier, V. Hypga, and P. Kopacek. 2003. Lysozyme from the gut of the soft tick, Ornithodoros moubata, the sequence, phylogeny and post-feeding regulation. Dev. Comp. Immunol. 27: 651Ð 660. Johns, R., D. E. Sonenshine, and W. L. Hynes. 2000. Responses of the tick, Dermacentor variabilis (Acari: Ixodidae) to hemocoelic inoculation of Borrelia burgdorferi (Spirochetales). J. Med. Entomol. 37: 265Ð270. Johns, R., J. Onishi, A. Broadwater, D. E. Sonenshine, A. deSilva, and W. L. Hynes. 2001a. Contrasts in tick innate immune responses to Borrelia burgdorferi challenge: immunotolerance in Ixodes scapularis versus immunocompetence in Dermacentor variabilis (Acari: Ixodidae). J. Med. Entomol. 38: 99 Ð107. Johns, R., D. E. Sonenshine, and W. L. Hynes. 2001b. IdentiÞcation of a defensin from the hemolymph of the American dog tick, Dermacentor variabilis. Insect Biochem. Mol. Biol. 31: 857Ð 865. Jones, L. D., C. R. Davies, G. M. Steele, and P. A. Nuttall. 1987. A novel mode of arbovirus transmission involving a nonviraemic host. Science (Wash. DC) 237: 775Ð777. Kocan, K. M., D. Stiller, W. L. Goff, P. L. Claypool, W. Edwards, S. A. Ewing, T. C. McGuire, J. A. Hair, and S. J. Barron. 1992. Development of Anaplasma marginale in male Dermacentor andersoni transferred from infected to susceptible cattle. Am. J. Vet. Res. 53: 499 Ð507. Kocan, K. M., and J. de la Fuente. 2003. Co-feeding studies of ticks infected with Anaplasma marginale. Vet. Parasitol. 112: 295Ð305. Kocan, K. M., J. de la Fuente, E. F. Blouin, and J. C. GarciaGarcia. 2004. Anaplasma marginale (Rickettsiales: Anaplasmataceae): recent advances in deÞning hostpathogen adaptations of a tick-borne rickettsia. Parasitology 129: S285ÐS300. Kollien, A. H., S. Fechner, P. J. Waniek, and G. A. Schaub. 2003. Isolation and characterization of a cDNA encoding for a lysozyme from the reduviid bug Triatoma infestans. Arch. Insect Biochem. Physiol. 53: 134 Ð145. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Lond.) 227: 680 Ð 685. Lehane, M. J., D. Wu, and S. M. Lehane. 1997. Midgut speciÞc immune molecules are produced by the blood-sucking insect Stomoxys calcitrans. Proc. Natl. Acad. Sci. U.S.A. 1997. 94: 11502Ð11507.

Vol. 42, no. 5

Lowenberger, C. A., S. Kamal, J. Chiles, S. Paskewitz, P. Bulet, J. A. Hoffman, and B. Christensen. 1999. MosquitoÐPlasmodium interactions in response to immune activation of the vector. Exp. Parasitol. 91: 59 Ð 69. Macaluso, K. R., D. E. Sonenshine, S. M. Ceraul, and A. F. Azad. 2001. Infection and transovarial transmission of rickettsiae in Dermacentor variabilis ticks acquired by artiÞcial feeding. Vector Borne Zoonotic Dis. 1: 45Ð53. Munderloh, U. G., E. F. Blouin, K. M. Kocan, N. L. Ge, W. Edwards, and T. J. Kurtti. 1996. Establishment of the tick (Acari: Ixodidae)-borne cattle pathogen Anaplasma marginale (Rickettsiales: Anaplasmataceae) in tick cell culture. J. Med. Entomol. 33: 656 Ð 664. Nakajima, Y., A. V. van Naters-Yasui, D. Taylor, and M. Yamakawa. 2001. Two isoforms of a member of the arthropod defensin family from the soft tick, Ornithodoros moubata (Acari: Ixodidae). Insect Biochem. Mol. Biol. 31: 747Ð751. Nakajima, Y., A. van der Goes van Naters-Yasui, D. Taylor, and M. Yamakawa. 2002. Antibacterial peptide defensin is involved in midgut immunity of the soft tick, Ornithodoros moubata. Insect Mol. Biol. 11: 611Ð 618. Nakajima, Y., K. Ogihara, D. Taylor, and M. Yamakawa. 2003. Antibacterial hemoglobin fragments from the midgut of the soft tick, Ornithodoros moubata (Acari: Argasidae). J. Med. Entomol. 40: 78 Ð 81. Palmer, G. H., S. D. Waghela, A. F. Barbet, W. C. Davis, and T. C. McGuire. 1987. Characterization of a neutralization-sensitive epitope on the Am 105 surface protein of Anaplasma marginale. Int. J. Parasitol. 17: 1279 Ð1285. Piesman, J., M. C. Dolan, T. R. Burkot, and M. E. Schriefer. 1998. Ability of three person-biting tick species from eastern United States to transmit Borrelia burgdorferi, pp. 206 Ð211. In L. Coons and M. Rothschild [eds], Proceedings, 2nd International Conference on Tick-Borne Pathogens at the Host-Vector Interface. A Global Perspective, 28 AugustÐ1 September 1995. The Conference, Kruger National Park, South Africa. Randolph, S. E., L. Gern, and P. A. Nuttall. 1996. Co-feeding ticks: epidemiological signiÞcance for tick-borne pathogen transmission. Parasitol. Today 12: 472Ð 479. Rau, U., and C. Hannoun. 1968. The use of a capillary-tube technique for artiÞcially feeding Argas reflexus reflexus ticks. Bull. World Health Organ. 39: 332Ð333. Rechav, Y., M. Zyzak, L. J. Fielden, and J. E. Childs. 1999. Comparison of methods for introducing and producing artiÞcial infection of ixodid ticks (Acari: Ixodidae) with Ehrlichia chaffeensis. J. Med. Entomol. 36: 414 Ð 419. Saliki, J. T., E. F. Blouin, S. J. Rodgers, and K. M. Kocan. 1998. Use of tick cell culture-derived Anaplasma marginale antigen in a competitive ELISA for serodiagnosis of anaplasmosis. Ann. N.Y. Acad. Sci. 849: 273Ð281. Sonenshine, D. E. 1993. Biology of ticks, vol. 2. Oxford University Press, New York. Sonenshine, D. E., S. M. Ceraul, W. E. Hynes, K. R. Macaluso, and A. F. Azad. 2002. Expression of defensin-like peptides in tick hemolymph and midgut in response to challenge with Borrelia burgdorferi, Escherichia coli and Bacillus subtilis. Exp. Appl. Acarol. 28: 127Ð134. Received 1 February 2005; accepted 14 June 2005.