Evaluation of Doxycycline-Laden Oral Bait and Topical Fipronil Delivered in a Single Bait Box to Control Ixodes scapularis (Acari: Ixodidae) and Reduce Borrelia burgdorferi and Anaplasma phagocytophilum Infection in Small Mammal Reservoirs and Host-Seeking Ticks

Abstract

A field trial was conducted on residential properties in a Lyme disease endemic area of New Jersey to determine the efficacy of Maxforce Tick Management System (TMS) bait boxes modified with doxycycline hyclate-laden bait to reduce the acarological risk of Lyme disease and the utility of galvanized steel shrouds to protect the bait boxes from squirrel depredation and ability to routinely service these devices. The strategy began with a 9-wk deployment against larvae followed by a 17-wk deployment against nymphs and larvae the second year. Passive application of fipronil reduced nymphal and larval tick burdens on small mammals by 76 and 77%, respectively, and nymphal tick abundance by 81% on treated properties. In addition, the percentage of infected small mammals recovered from intervention areas following treatment was reduced by 96% for Borrelia burgdorferi and 93% for Anaplasma phagocytophilum. Infection prevalence in host-seeking nymphal ticks for both B. burgdorferi and A. phagocytophilum were reduced by 93 and 61%, respectively. Results indicate that Maxforce TMS bait boxes fitted with doxycycline-impregnated bait is an effective means of reducing ticks and infection prevalence for B. burgdorferi and A. phagocytophilum in both rodent reservoirs and questing Ixodes scapularis Say ticks. The protective shroud allows the device to be routinely serviced and protect against squirrel depredation.

The incidence of disease caused by Ixodes scapularis Say-transmitted pathogens continues to escalate in many parts of the United States and as a consequence, the development of reliable means to reduce the risk of pathogen transmission remains a public health priority (Hayes and Piesman 2003, Piesman and Eisen 2008, Centers for Disease Control and Prevention 2015). Reduction of vector tick populations is among the most effective ways to mitigate disease transmission risk (Hayes et al. 1999, Hayes and Piesman 2003), and the application of chemical acaricides to tick habitat is perhaps the most widely investigated, rapid, and reliable means of suppressing I. scapularis (Stafford and Kitron 2002; Schulze et al. 2005a, 2008). Public perception, however, generally views the use of habitat-targeted synthetic acaricides as potentially having adverse health effects and undesirable environmental impacts (Ginsberg 1994, Schmidtmann 1994, Schulze et al. 2001b, Gould et al. 2008). As a result, < 25% of homeowners residing in Lyme disease-endemic communities report treating their properties to control ticks (Piesman 2006, Schulze et al. 2007).

Host-targeted chemical control provides an alternative to area-wide or barrier applications of acaricide. Topical application of acaricide to reduce infestation by I. scapularis adults parasitizing white-tailed deer (Odocoileus virginiana Zimmerman) or loads of immature ticks on small mammals has been shown effective when used alone or in combination as part of an overall integrated approach (Mather et al. 1987, Solberg et al. 2003, Dolan et al. 2004, Schulze et al. 2007, Brie et al. 2009). Although these technologies have the advantage of minimizing the amount of acaricide introduced into the environment, their widespread use has been hampered by regulatory constraints and relatively high cost (Schulze et al. 2007, Piesman and Eisen 2008). Recently, the use of doxycycline hyclate to target tick-borne pathogens within vertebrate hosts has been proposed (Zeidner et al. 2004, Massung et al. 2005, Dolan et al. 2008) and shown in field trials to be effective in prophylactically protecting small mammal reservoirs, curing infected reservoirs, and clearing I. scapularis feeding on treated reservoirs of the Lyme disease spirochete, Borrelia burgdorferi. Dolan et al. (2011) reported that 500 mg/kg doxycycline hyclate-laden bait reduced B. burgdorferi infection rates in small mammal reservoirs and questing nymphal ticks by 87 and 94%, respectively. In this study, we examine the efficacy of doxycycline hyclate-laden baits when deployed together with topically applied fipronil in a rodent bait box to both suppress tick populations and reduce infection in questing ticks and their small mammal hosts; and, as an adjunct, explore the possibility of the efficient and cost-effective reuse of the bait boxes when fitted with protective galvanized steel shrouds.

Materials and Methods

Study Areas

The trial was conducted on four residential properties located in Millstone Township, Monmouth County, NJ, described previously (Schulze et al. 2007). Each of the ≥1.0-ha properties were situated within oak-dominated forests and consisted of lawn-landscaping immediately around the residence surrounded by woodland (±60–70% of each property). Three similar residential properties and a nearby natural area served as control sites. Earlier studies conducted throughout this area showed I. scapularis and its small mammal hosts to be abundant (Schulze et al. 2001c, 2005b, 2007).

Bait Box and Modifications

The Maxforce Tick Management System (TMS) is a 19.05- by 13.97- by 6.35-cm child-resistant, injection-molded plastic bait box. After entering the box, target small mammals are passively treated as they contact a fipronil-treated felt wick while attempting to access an attractive bait. Treatment after a single visit to a bait box was effective in killing ticks on small mammals for up to 42 d (Dolan et al. 2004).

Three major modifications were made to the Maxforce TMS bait boxes. First, we removed the secured lid by removing the four drive screws holding it in place. Second, we substituted 0.05% doxycycline hyclate-impregnated baits (Genesis Laboratories, Inc., Wellington, CO) for the as-manufactured bait blocks and replaced the lid. Finally, because bait boxes deployed in an earlier study were commonly damaged by eastern gray squirrels (Sciurus carolinensis Gmelin) (Schulze et al. 2007), we installed a two-piece, tightly fitting protective cover constructed of 0.032 gauge galvanized steel developed by Connecticut Tick Control (Norwalk, CT). The top and bottom sections of the clam-shell cover and bait box were secured together at two opposite corners using 20.3-cm plastic cable ties (Thomas & Betts Corp., Memphis, TN).

Bait Box Deployment

Original label directions for MaxForce TMS suggest a deployment in May targeting nymphal ticks followed by a second deployment in late July/early August targeting larvae. This deployment strategy was demonstrated in a previous study having reported an initial deployment in May against nymphs followed by a second deployment in late July for larvae (Schulze et al. 2007). In this study, 24 modified bait boxes were deployed at each of the four treatment properties in late July 2008 for 9 wk. We chose to make our initial deployment in late July against larvae because doxycycline-laden baits were previously shown to clear infection in small mammals (Dolan et al. 2011) and we expected that larvae questing in the treated areas would be likely to obtain a bloodmeal from pathogen-free animals. In addition, the inclusion of the fipronil wick would kill ticks on any treated mice. We thus expected fewer larvae should feed and molt to nymphs and that fewer of the subsequently emerging nymphs would carry infection. In 2009, bait boxes were deployed for 17 wk between mid-May and mid-September. Bait boxes were deployed at ∼25-m intervals along two concentric rings, with the first located within the forest at ± 10 m from the lawn edge within surrounding forest. Wherever possible, boxes were placed in proximity to likely small mammal activity such as woodpiles, brush piles, fallen logs, etc.

Bait Box Maintenance

Bait boxes were visited weekly and weighed in the field using a Scout Pro Balance (Ohaus Corporation, Pine Brook, NJ). Boxes exhibiting a loss in weight of 5.0 g were considered used. Boxes with a reduction in weight of ≥ 50.0 g, indicating ≈ 50% of bait had been removed (Schulze et al. 2007), were opened, re-baited, resealed as described above, and reweighed. Any change in box orientation or damage to boxes, including wicks, was noted and damaged or missing wicks were immediately replaced. All wicks were retreated with 0.70% fipronil in late July 2009 prior to the larval season.

Small Mammal Trapping and Tick Burdens

Small mammals were collected from treatment properties and from control properties and natural areas using 7.6- by 8.9- by 30.5-cm Sherman nonfolding box traps (H.B. Sherman, Tallahassee, FL) baited with rolled oats and cotton balls. In 2008, preintervention trapping was conducted in mid-May and mid-July during the activity periods of I. scapularis nymphs and larvae, respectively, while postintervention trapping was conducted in late August, to assess larval tick burdens (Schulze et al. 1986, 2005b, 2007). In 2009, preintervention trapping was conducted May 18–21, while postdeployment trapping was performed between June 15–17, and August 17–19. Live captured mice and chipmunks were used for all analysis (incidental captures of short-tailed shrews and eastern gray squirrels were excluded). During each trapping event, 25 Sherman traps were set at each residential treatment and control site, while 100 Sherman traps were deployed at the natural control site. Traps were deployed by mid-afternoon and checked by mid-morning the following day. Traps remained open during the day and checked periodically until late afternoon. Captured rodents were transported to a central location, anesthetized with isoflurane, and processed as follows: collection of blood and ear biopsy samples, examination for ticks, recording of physical measurements, and marking with individual metal ear tags (Monel Model 1005-1 or 1005-3, National Band and Tag Company, Newport, KY). Captures were then placed back into traps, allowed to recover from the anesthetic, and released at the point of capture. Ear tissue biopsies, as described by Sinsky and Piesman (1989), were surface sterilized and placed in BSK-H media for culture of live spirochetes and cultures were read by dark-field microscopy every 7 d for 4 wk before being deemed negative. Cultures were used to determine infection rates in small mammals for B. burgdorferi. Blood samples recovered from small mammals were evaluated by PCR to determine infection rate for A. phagocytophilum. Ticks collected from small mammals were placed in discrete vials containing 70% ethanol and labeled with the corresponding ear tag and collection numbers. Small mammals recaptured during a particular trapping event were not reprocessed.

Collection of Host-Seeking Ticks

Host-seeking adult and nymphal I. scapularis were collected using a combination of dragging and walking methods (Ginsberg and Ewing 1989, Schulze et al. 1997). For each sampling event, ticks were collected from five 100-m transects located in suitable habitat at each site and placed in discrete vials containing 70% ethanol for subsequent analysis. If necessary, sampling continued until a minimum of 25 ticks was collected from each property, yielding a combined total of 100 ticks each from treatment and control sites. Pre-(2008) and post- (2009–2010) intervention sampling for I. scapularis nymphs was performed to coincide with their respective trapping events in 2009, while final collections were made in June 2010. Adult I. scapularis were sampled in early November 2008 and 2009 and April 2009 and 2010. All sampling was performed between 0800–1200 h when vegetation was dry and wind was below 10 km/h (Schulze et al. 2001a, Schulze and Jordan 2003).

Analysis of Ticks and Small-Mammal Blood for B. burgdorferi and Anaplasma phagocytophilum Infection

DNA was isolated from ticks using DNeasy Blood and Tissue kit (Qiagen, USA) and a Mini-Beadbeater (Biospec, USA) as described previously (Dolan et al. 2011). The protocol was a modification of Crowder et al. (2010); one tick was added to a 0.5-ml free standing screw cap tube (Fisher Scientific, USA) with 450 µl ATL buffer (Qiagen, USA), 20 µl proteinase K, 10 µg Polyadenylic Acid (Amersham Bioscience, USA), 400 mg 2.0 mm zirconia beads (Biospec), 90 mg 0.1 mm zirconia/silica beads (Biospec), processed with the Mini-Beater (Biospec) for 4 min, and incubated for 15 min at 56°C. Following incubation, the sample was centrifuged at 6,000× g for 1 min, and 425 µl of the sample was mixed with 425 µl AL buffer (Qiagen, USA) and incubated at 70°C for 10 min. After incubation, the sample was centrifuged at 20,000× g for 20 s and processed using the DNeasy Blood and Tissue Kit (Qiagen), following the animal tissue protocol starting at step 4. A vacuum manifold was used instead of centrifugation steps. The final DNA was eluted with 50 µl of buffer AE (Qiagen, USA).

Three separate real-time PCR (qPCR) assays were performed in 25 µl solutions, each with 12.5 µl iQ Multiplex Powermix (BioRad, USA), 5 µl tick extract, forward and reverse primers in a final concentration of 300 nM each, and probes in a final concentration of 200 nM. The PCR cycling conditions for all assays consisted of denatured DNA at 95°C for 3 min followed by 40 cycles of 95°C for 10 s, and 60°C for 1 min on a C1000 Touch thermal cycler with a CFX96 real time system (BioRad). Water was used as a negative control on each 96 well plate.

To ensure the presence of tick DNA, all tick extracts were examined by qPCR for the actin locus. The primers used for the actin control PCR were actin-F 5’-GCCCTGGACTCCGAGCAG-3’ and actin-R 5’-CCGTCGGGAAGCTCGTAGG-3’. The probe used was actin-probe Quas705-CCACCGCCGCCTCCTCTTCTTCC-BHQ3.

To verify the presence of B. burgdorferi DNA in a sample, a second qPCR was performed amplifying the fliD locus. The primers used in this TaqMan assay were fliD-F 5’-TGGTGACAGAGTGTA TGATAATGGAA-3’, fliD-R 5’-ACTCCTCCGGAAGCCACAA-3’, and the probe used was fliD-probe FAM-TGCTAAAATGCTAGGA GATTGTCTGTCGCC-BHQ1, which were described previously by Zeidner et al. (2001).

A third qPCR was performed to verify the presence of A. phagocytophilum DNA amplifying the msp2 locus. The primers for this qPCR assay were msp2-F ATGGAAGGTAGTGTTGGTTATGGTA TT and msp2-R TTGGTCTTGAAGCGCTCGTA, and the probe used was msp2-probe HEX-TGGTGCCAGGGTTGAGCTTGAG ATTG-BHQ1 (Zeidner et al. 2000, Massung and Slater 2003).

DNA was extracted from small-mammal blood samples (50 µl blood pellet per animal) using the Qiagen DNeasy blood and tissue kit (Qiagen) following the “Purification of Total DNA from Animal Blood or Cells” protocol, with a final elution volume of 200 µl per sample as described previously (Dolan et al. 2011). Two qPCR assays were run on each mammal blood sample, following the same protocols above for the fliD and msp2 loci.

Analysis of Small-Mammal Blood for Doxycycline Hyclate

High-performance liquid chromatography (HPLC) was performed to determine plasma pharmacokinetic levels for small mammals as previously described (Zeidner et al. 2004, Dolan et al. 2008, 2011). Approximately 200 µl whole blood were collected in microtainers containing EDTA and centrifuged, and the plasma was run over a Beckman System Gold high-pressure liquid chromatograph (Beckman Coulter, Fullerton, CA); associated 32-Karat version 5.0 software was used combined with a C₁₈ column (100 by 4.6-mm inside diameter; Alltech, Deerfield, IL).

Statistical Analyses

The modified Abbott’s formula (Mount 1981) was used for primary comparisons between untreated control and treatment areas, including percent reduction of B. burgdorferi and A. phagocytophilum infection rates in small mammals, infestation rates of small mammals, and questing larval and nymphal ticks. ANOVA and Mann Whiney U tests were performed to determine any significance in nymphal infestation rates and tick burdens among small mammals and density of host-seeking ticks between treated and untreated areas, respectively. Percent control was calculated using the modified Henderson’s equation as follows: 100-(treatment/untreated control × 100). Chi Square (χ²) tests were used to determine significance (at P < 0.05).

Results

Squirrel Depredation

During 6,048 deployment days (96 boxes deployed for 63 d) in 2008, no squirrel depredation of the cover-equipped bait boxes was observed. Sixteen boxes were found inverted (which does not affect performance), and damaged or missing wicks were replaced in a total of 41 boxes. In 11,424 deployment days (96 boxes deployed for 119 d) in 2009, 85 boxes were found inverted and wicks were replaced in a total of 49 boxes. Between weeks 4 and 12, we observed damage to one or both bait chambers in 20 boxes at one of the treatment sites. Damage was confined to the rear wall of the bait chambers nearest the box entry portals and did not affect the integrity of the wicks or product label. All boxes were accounted for at the conclusion of the trial.

Bait Consumption

After the first week of deployment in 2008, > 65% of the bait boxes demonstrated a decline in weight ≥ 5.0 g, indicating use (Schulze et al. 2007), a weekly trend that continued throughout the trial. Bait consumption ≥ 5.0 g was observed in > 60% of deployed boxes for 6 of the 9 wk, peaking at 71% at 6 wk. Over the entire deployment period, all bait was removed from an average of 44% (range for the four treated properties = 19–62%) of boxes within 1 wk of deployment. Overall bait consumption began to decline after 6 wk of deployment. In 2009, 79% of boxes were used by the end of the second week and > 90% of boxes showed activity between weeks 3–12, dropping to a low of 75% by the conclusion of the deployment. Between weeks 4–10, > 87% of boxes had to be re-baited, but the percentage declined gradually thereafter with the appearance of oak mast. In 2009, all bait was removed from an average of 69% (range for the four treated properties = 15–92%) of boxes.

Bait Box Maintenance

To service the bait boxes, the upper section of the protective cover was removed by cutting the cable ties. New baits and, as needed, a new wick assembly were installed, the box lid reinstalled, and the protective cover secured with cable ties. With some experience, a bait box could be serviced and reweighed in under a minute. Excluding travel, the typical weekly maintenance of bait boxes required about 30 min per property.

Small Mammal Trapping and Tick Burdens

We captured a total of 401 small mammals (215 in treatment sites and 186 in control sites). White-footed mice were 55.9% of captures at the untreated areas, whereas eastern chipmunks (Tamias striatus L.) comprised 63.7% of captures at the treatment sites. Northern short-tailed shrews (Blarina brevicauda Say) and eastern gray squirrels (Sciurus carolinensis Gmelin) made up 6.9% of total captures.

Prior to bait box deployment in 2008, May–June nymphal I. scapularis infestation rates (46.2%) and mean nymphal tick burdens (1.7 ± 0.3 nymphs per animal) in treatment areas were higher but did not differ significantly from those at the control sites [38.8% (χ²₍₂₎ = 0.46; P = 0.49) and 1.0 ± 0.2 nymphs per animal (one-way ANOVA: F_(1,130)= 0.59; P = 0.44)] (Table 1). Infestation rates and mean tick burdens declined during each sampling period after deployment (Table 1). Similarly, preintervention small mammal trapping in July 2008 showed no significant difference in I. scapularis larval infestation rates (χ²₍₂₎ = 0.16; P = 0.69) nor larval tick burden burdens (one-way ANOVA: F_(1,51)= 0.05; P = 0.82) between treated properties and control areas (Table 2). However, infestation rates and mean tick burdens in the treated areas declined substantially after intervention. While mean tick burdens more than doubled between years at untreated areas, nymphal and larval infestation rates declined by 71.7% and 80.8%, respectively, in 2009; while tick burdens declined by 87.2% and 94.4% for nymphs and larvae, respectively (Table 2).

Table 1.

Ixodes scapularis burdens on live-captured small mammals in control and treatment sites before intervention (May–June 2008) and after intervention with rodent bait boxes equipped with fipronil-treated wicks and doxycycline hyclate-laden oral bait (August 2008 to August 2009)

Collection month	Sites	Infestation ratea	% Reductionb	Mean ticks/animalc	% Reductiond
2008
May/June	Control	26/67 (38.8%)		1.0 ± 0.2 (70/67)
	Treatmente	30/65 (46.2%)		1.4 ± 0.3 (89/65)
July	Control	15/25 (60.0%)		2.7 ± 0.9 (67/25)
	Treatmentf	18/26 (69.2%)		2.9 ± 0.7 (76/26)
Intervention
August	Control	18/31 (58.1%)		2.6 ± 0.7 (80/31)
	Treatment	5/45 (11.1%)	80.9%	0.2 ± 0.1 (10/45)	92.3%
2009
May	Control	20/21 (95.2%)		5.8 ± 1.0 (122/21)
	Treatment	13/22 (59.1%)	37.9%	2.0 ± 0.6 (44/22)	65.5%
June	Control	27/34 (79.4%)		3.9 ± 0.7 (131/34)
	Treatment	9/40 (22.5%)	71.7%	0.5 ± 0.2 (18/40)	87.2%
August	Control	16/20 (80.0%)		10.8 ± 3.5 (216/20)
	Treatment	4/26 (15.4%)	80.8%	0.6 ± 0.3 (16/26)	94.4%

^aInfestation rate is represented as total number of animals with ticks/total number of animals examined (% infested animals).

^b% Reduction in infestation rate = 100 × [1– (infection rate in treated area/infection rate in untreated area)], after Abbott (1925).

^cMean (± SE) ticks/animal is represented as mean number of ticks (total number of ticks/total number of animals examined).

^d% Reduction in mean ticks/animal = 100 × [1– (mean ticks/animal in treated area/mean ticks/animal in untreated area)], after Abbott (1925).

^eNeither nymphal infestation rate (χ²₍₂₎ = 0.46; P = 0.49) nor nymphal tick burden (one-way ANOVA: F_(1,130)= 0.59; P = 0.44) differed between treatment and control areas prior to intervention.

^fNeither larval infestation rate (χ²₍₂₎ = 0.16; P = 0.69) nor larval tick burden (one-way ANOVA: F_(1,51)= 0.05; P = 0.82) differed between treatment and control areas prior to intervention.

Table 2.

Density of host-seeking I. scapularis nymphs and adults in control and treatment sites, May 2008–2010

Sites	Nymphs (May–June)			Adults (Sept.–Oct.)
	2008	2009	2010	2009	2010
Controla,b	8.3 ± 0.6	9.0 ± 1.0	8.5 ± 0.9	7.0 ± 1.7	6.3 ± 2.2
Treatment	7.3 ± 1.3	2.5 ± 0.3 (68.4%)c	1.7 ± 0.3 (77.2%)c	0.8 ± 0.3 (88.6%)d	0.4 ± 0.1 (93.7%)d

^aValues are ticks collected/100-m drag (mean ± SE for n = 5 100-m transects for each site).

^bNo significant difference in density of host-seeking I. scapularis larvae (Mann–Whitney U_(10,10) = 44.0: P = 0.65) between treatment and control areas prior to intervention.

^cPercent control, after Henderson’s equation: percent control = 100 − (T/U × 100), where T and U are the mean after treatment/mean before treatment in treated plots and untreated plots, respectively.

^dPercent reduction = 100 × [1– (mean ticks/transect in treated area)/(mean ticks/transect in untreated area)], after Abbott (1925).

Abundance of Host-Seeking Ticks

We found no significant difference in the density of host-seeking I. scapularis (Mann–Whitney U_(10,10) = 44.0; P = 0.65) larvae between treatment and control areas prior to bait box deployment (Table 2). Deployment of bait boxes against I. scapularis larvae in August 2008 and continuing in 2009 resulted in 72.2 and 81.1% control of host-seeking nymphs and 88.6 and 93.7% control of adults in 2009 and 2010, respectively, compared to control sites (Table 2).

Infection Rates in Small Mammals

Predeployment (May–June 2008) B. burgdorferi infection rates in small mammals, based on culture of ear biopsies, averaged 62.1% and 49.3% for treated and untreated areas, respectively (Table 3). After deployment, only 5.3% (8/151) of small mammals from treated areas tested yielded culture positive ear biopsies in months postdeployment as compared with 56.6% (64/113) for control areas. Predeployment A. phagocytophilum infection rates in small mammals from the treated and untreated areas were 37.9 and 41.1%, respectively (Table 3). In August of 2008 and June of 2009, the average rate of infection in small mammals in treated areas was reduced to 5.6% (6/108), while no infected animals [0/19 (0.0%)] were captured in August of 2009.

Table 3.

Infection with B. burgdorferi and A. phagocytophilum in small mammals in treatment and control sites before intervention (May–June 2008) and after intervention with rodent bait boxes equipped with fipronil-treated wicks and doxycycline hyclate-laden oral bait (August 2008 to August 2009)

Date	Doxycycline blood level µg/ml		B. burgdorferi infection rate		A. phagocytophilum infection rate
	Treatment	Control	Treatment	Control	Treatment	Control
2008
May–June	NAa	NA	41/66 (62.1%)	37/75 (49.3%)	25/66 (37.9%)	23/56 (41.1%)
August	1.78	0.025b	1/50 (2.0%)	15/34 (44.1%)	1/35 (2.9%)	3/8 (37.5%)
2009
May	NAa	NAc	5/24 (20.8%)	12/22 (54.6%)	4/47 (8.5%)	5/13 (38.5%)
June	1.13	0.0	1/47 (2.1%)	23/36 (63.9%)	1/26 (3.9%)	ND
August	1.44	0.0	1/30 (3.3%)	14/21 (66.7%)	0/19 (0.0%)	ND

^aMinimum inhibitory concentration (MIC) for doxycycline blood levels is ∼0.5 µg/ml (Zeidner et al. 2004, Dolan et al. 2008; 2011).

^bAlthough there was no doxycycline bait deployed in the untreated areas, one eastern chipmunk had detectable doxycycline blood levels as determined by HPLC (0.87 µg/ml). This animal most likely entered from one of the two control areas located ≥ 500 m from the treatment area.

^cMay of 2008 and 2009 were annual pretreatment trapping events and no doxycycline bait was deployed during these periods; therefore, blood was not tested by HPLC. NA = not applicable.

Infection Rates in Questing Ticks

Initial (2008) average infection rates for both B. burgdorferi and A. phagocytophilum in questing ticks declined slightly in the untreated areas in 2010 (Table 4). In treated areas, 24.6% of nymphs and 51.6% of adult were infected with B. burgdorferi during the predeployment period, whereas infection rates declined to 1.8 and 10.4% representing a 90.6 and 74.8% reduction, respectively. We also saw a substantial reduction in A. phagocytophilum infection in questing nymphal (53.2%) and adult (58.4%) ticks.

Table 4.

Infection rate with B. burgdorferi and A. phagocytophilum in questing nymphal and adult I. scapularis in treatment and control sites for time periods before the intervention could impact the infection rate in questing nymphs or adults (2008) and after intervention with rodent bait boxes equipped with fipronil-treated wicks and doxycycline hyclate-laden oral bait (2009–2010)

Year	Agent	Infection ratea (Treatment)		Infection ratea (Control)		% Reductionb
		Nymphs	Adults	Nymphs	Adults	Nymphs	Adults
2008	Bbc	30/122 (24.6%)	51/100 (51.6%)	15/56 (26.8%)	49/93 (52.7%)	–	–
	Apd	23/122 (18.9%)	19/100 (19.0%)	11/56 (19.6%)	13/59 (22.0%)	–	–
2009	Bb	7/58 (12.1%)	12/42 (28.6%)	12/42 (28.6%)	29/55 (52.7%)	57.7%	45.73%
	Ap	8/58 (13.8%)	4/42 (9.5%)	7/42 (16.7%)	12/55 (21.8%)	17.4%	56.42%
2010	Bb	1/56 (1.8%)	6/58 (10.4%)	23/120 (19.2%)	37/85 (43.5%)	90.63%	74.83%
	Ap	4/56 (7.4%)	5/58 (8.6%)	18/120 (15.8%)	17/85 (20.0%)	53.16%	58.38%

^aInfection rate in both treatment and control areas for B. burgdorferi and A. phagocytophilum in nymphal and adult I. scapularis is represented as number of positive ticks/total number of ticks tested (% infected) as determined by PCR.

^b% Reduction for B. burgdorferi and A. phagocytophilum infection in nymhpal and adult I. scapularis ticks is a comparison of ticks collected in treated and untreated control areas using the modified Henderson’s equation; percent reduction = 100 – (T/U × 100), where T and U are the mean after treatment/mean before treatment in treated plots and untreated plots, respectively.

^cBb = Borrelia burgdorferi.

^dAp = Anaplasma phagocytophilum.

Discussion

Our results demonstrated that modified bait boxes reduced larval and nymphal I. scapularis burdens on small mammal hosts, host-seeking tick populations, and B. burgdorferi and A. phagocytophilum infection rates in questing ticks following a ∼26-wk deployment. In a previous study, prototype bait boxes reduced nymphal and larval I. scapularis burdens on mice by 68 and 84%, respectively. As a result, nymphal and adult questing tick populations were reduced by >50 and 77%, respectively (Dolan et al. 2004). Additionally, reduction in B. burgdorferi infected nymphs decreased overall reservoir capacity and afforded young mice protection from infected nymphal ticks resulting in a >57% reduction of infected reservoir mice. Although we are unable to determine the relative contribution made by fipronil and doxycycline bait in this study, their combined use in this study resulted in a substantially higher reduction in reservoir hosts than that achieved in the previous trial using fipronil alone. However, based on previous studies, we believe there was a combined effect with fipronil most likely having a lesser impact in comparison to the doxycycline bait (Dolan et al. 2004, 2011).

In a previous study, we distributed the same doxycycline baits used in this study in rodent-targeted bait boxes during the nymphal and larval seasons over a 2-yr period and demonstrated a significant impact on the natural enzootic cycles of both B. burgdorferi and A. phagocytophilum in small mammals and I. scapularis ticks (Dolan et al. 2011). The percentage of small mammals infected with B. burgdorferi and A. phagocytophilum were reduced by 87 and 74%, respectively. In addition, infection rates in host-seeking nymphal ticks were reduced by 94% for B. burgdorferi and 92% for A. phagocytophilum. We obtained similar results over a shorter period of intervention that may represent an additive effect of using fipronil and antibiotic baits concurrently.

Consumption of doxycycline bait effectively cleared small mammal reservoir hosts (primarily white-footed mice and Eastern chipmunks) of infection in the treated areas. By initiating treatment in July 2008 during the larval tick season, host-seeking larvae had a much greater chance of acquiring an uninfected host that had also been treated with fipronil. As a result, fewer nymphs emerged during spring following each year of treatment and those that did demonstrated a marked reduction in infection rates for both pathogens. Subsequent treatment during the following spring compounded the effect against questing nymphs that escaped treatment as larvae, with consequent effects on numbers of infected adults.

Addition of the protective cover did not affect small mammal use of the bait boxes and provided near complete protection of bait boxes from damage by eastern gray squirrels. Rates of bait consumption suggested that bait boxes and doxycycline bait were readily used by targeted small mammals. The decline in bait consumption observed in fall 2009 coincided with the appearance of an alternative food source (substantial oak mast) previously shown to affect efficacy of host-targeted strategies (Ostfeld et al. 2006). Although bait consumption declined somewhat after 10 wk in apparent response to heavy oak mast in fall of 2009, bait box use remained high with an average of >84% of boxes showing use. Only 20 bait boxes were damaged in 17,472 deployment days, while in four earlier deployments of 9,800 days each, up to 92% of unprotected bait boxes were damaged by squirrels, with a large portion of the boxes damaged after only 1 wk (Schulze et al. 2007, T. L. Schulze, unpublished data).

The protective cover also allowed us to efficiently service the bait boxes in the field. The two-piece cover could be quickly removed by cutting the cable ties, facilitating replacement of baits and retreating or replacement of wicks in situ. Although the product labeling recommends separate deployments of bait boxes against I. scapularis nymphs and larvae, our results demonstrated that bait boxes, fitted with a secured protective cover, could be left in the field for extended periods and serviced periodically to meet the needs of the user. We expect that labor costs incurred for servicing the bait boxes would be offset by eliminating labor costs for retrieval and redeployment of a second set of boxes. Also, because the original Maxforce TMS box and currently available SELECT TCS bait boxes (Tick Box Technology Corporation, Norwalk, CT) are intended to be discarded after a single 3-mo deployment, maintenance costs (bait, acaricide, replacement wick assemblies) could be further offset by elimination of replacement costs, as well as the financial and environmental costs of disposal. The current EPA-registered SELECT TCS bait boxes are essentially identical in technology and design. The one difference in the recently registered (2011) boxes is they are equipped with a galvanized metal shroud to ensure child-resistant packaging by preventing depredation by squirrels and raccoons.

Attempting to control ticks by integrating multiple modalities remains in early research stages. However, studies suggest that an integrated approach targeting multiple embryonic stages may offer superior levels of tick control as compared to a single approach (Stafford and Kitron 2002, Schulze et al. 2007), and deficiencies in one control method may be overcome by a concurrent intervention. Schulze et al. (2007) reported that the combined use of 4-Poster topical treatment devices for deer, Maxforce TMS bait boxes, and barrier applications of granular acaricide that sequentially targeted all active stages of I. scapularis, provided superior levels of tick control when compared to use of each method alone. We employed a single host-targeted device that included dual control modalities, fipronil to kill immature ticks and doxycycline bait to prophylactically protect and clear small mammals of infection as well as reduce infection rates in questing tick population that may remain following treatment. We believe that our results demonstrate the ability to effectively reduce tick burdens on small mammals, number of host-seeking ticks, and infection rates in rodent reservoirs and ticks, thus potentially interrupting the natural enzootic transmission cycle of B. burgdorferi among I. scapularis and its small mammal hosts.

We also demonstrated modification of a host-targeted technology to effectively provide multiple interventions using a single device while dramatically reducing the amount of acaricide introduced into the environment compared to traditional area-wide applications. Although additional research is needed to refine the approach, this technology offers an alternative for reducing the abundance of infected ticks in residential landscapes. The distribution of doxycycline-impregnated bait for controlling tick-borne diseases may be controversial given the fact that doxycycline is routinely prescribed for treating these infections in patients. However, we believe that this and previous studies support consideration of bait containing doxycycline or other antibiotics that do not serve as front-line drugs for treatment of human or veterinary diseases as part of an integrated-tick-management approach for the reduction of tick-borne diseases.