Ehrlichia and Spotted Fever Group Rickettsiae Surveillance in Amblyomma americanum in Virginia Through Use of a Novel Six-Plex Real-Time PCR Assay

Abstract

The population of the lone star tick Amblyomma americanum has expanded in North America over the last several decades. It is known to be an aggressive and nondiscriminatory biter and is by far the most common human-biting tick encountered in Virginia. Few studies of human pathogen prevalence in ticks have been conducted in our state since the mid-twentieth century. We developed a six-plex real-time PCR assay to detect three Ehrlichia species (E. chaffeensis, E. ewingii, and Panola Mountain Ehrlichia) and three spotted fever group Rickettsiae (SFGR; R. amblyommii, R. parkeri, and R. rickettsii) and used it to test A. americanum from around the state. Our studies revealed a presence of all three Ehrlichia species (0–24.5%) and a high prevalence (50–80%) of R. amblyommii, a presumptively nonpathogenic SFGR, in all regions surveyed. R. parkeri, previously only detected in Virginia's Amblyomma maculatum ticks, was found in A. americanum in several surveyed areas within two regions having established A. maculatum populations. R. rickettsii was not found in any sample tested. Our study provides the first state-wide screening of A. americanum ticks in recent history and indicates that human exposure to R. amblyommii and to Ehrlichiae may be common. The high prevalence of R. amblyommii, serological cross-reactivity of all SFGR members, and the apparent rarity of R. rickettsii in human biting ticks across the eastern United States suggest that clinical cases of tick-borne disease, including ehrlichiosis, may be commonly misdiagnosed as Rocky Mountain spotted fever, and that suspicion of other SFGR as well as Ehrlichia should be increased. These data may be of relevance to other regions where A. americanum is prevalent.

Introduction

Virginia is a state with diverse and changing tick-borne disease epidemiology that includes human ehrlichiosis, Rocky Mountain spotted fever (RMSF), Tidewater spotted fever (an illness caused by Rickettsia parkeri; Wright et al. 2011), Lyme disease, and anaplasmosis. The lone star tick Amblyomma americanum, the most common tick to bite people in the southeastern United States, is increasingly a cause of disease transmission (Paddock and Yabsley 2007, Apperson et al. 2008, Smith et al. 2010, Stromdahl and Hickling 2012), and is likely among the most important arthropod vectors of disease to humans in Virginia.

A. americanum potentially transmits ehrlichiosis and rickettsiosis. The case rates of these diseases vary across Virginia (Fig. 1). Human ehrlichiosis attributed to Ehrlichia chaffeensis may cause a mild to fatal illnesses, with a hospitalization rate of 49% and a fatality rate of 1.9% (Dahlgren et al. 2011). Ehrlichiosis caused by E. ewingii is less commonly reported and tends to be associated with previously immune-suppressed patients (Paddock and Yabsley 2007). In 2006, a new Ehrlichia species was discovered in A. americanum in Georgia (Loftis et al. 2006). This new species, known as Panola Mountain Ehrlichia (PME), was subsequently associated with human illness (Reeves et al. 2008) and has been detected in A. americanum collected from 10 different states, including Virginia (Loftis et al. 2008, Yabsley et al. 2008).

FIG. 1.

Mean number of human cases of ehrlichiosis or spotted fever rickettsiosis in Virginia counties and cities averaged over a 3-year period from 2009–2011.

During the past decade, a number of state health departments in the southeastern United States, including the Virginia Department of Health (VDH), have seen large increases in diagnosed and reported cases of RMSF despite a decline in the mortality rate (i.e., a rise in national incidence from 1.7 to 7 cases per million but decrease in mortality from 2.2% to 0.3%) (Chapman et al. 2006, Openshaw et al. 2010, Virginia Department of Health 2012). The agent of RMSF is Rickettsia rickettsii, and Dermacentor variabilis, the American dog tick, is thought to be the primary vector. However, R. rickettsii has rarely been found in the thousands of D. variabilis tested over the past two decades (Ammerman et al. 2004, Dergousoff et al. 2009, Moncayo et al. 2010, Stromdahl et al. 2011). This leaves open the possibility that A. americanum may play a role as a vector (Stromdahl et al. 2011).

A. americanum is also recognized as an important host of Rickettsia amblyommii, a spotted fever group Rickettsia (SFGR) with no apparent pathogenicity for humans. Mixson et al. (2006) found an R. amblyommii infection rate in adult A. americanum of 41.2% from nine states across the eastern United States. Several recent tick surveys in Virginia also found R. amblyommii to be common in A. americanum, with rates of 26–70% (J.S. and H.G., unpublished data) (Jiang et al. 2010).

Another SFGR potentially transmitted by A. americanum is R. parkeri. Although identified over 50 years ago (Lackman et al. 1949, Lackman et al. 1965), R. parkeri was only recognized as a human pathogen in Virginia in 2002 (Paddock et al. 2004). Since that time, human illness caused by R. parkeri has been identified across the southeastern United States (Paddock et al. 2008). The primary vector of R. parkeri is Amblyomma maculatum, the Gulf Coast tick, with high infection rates (41–43%) in ticks collected in Virginia (Fornadel et al. 2011,Wright et al. 2011). However, R. parkeri has also been detected in adult A. americanum from Tennessee and Georgia (Cohen et al. 2009). R. parkeri can infect A. americanum, be transmitted transstadially and transovarially, and subsequently be transmitted to guinea pigs by feeding A. americanum (Goddard 2003).

Few studies of tick species distribution and human pathogen prevalence have been conducted in Virginia since the mid-twentieth century, when research focused on D. variabilis and R. rickettsii (Sonenshine et al. 1966) and more recent surveys have focused on ticks other than A. americanum (Fornadel et al. 2011, Nadolny et al. 2011, Wright et al. 2011). In the intervening decades, deer populations have increased and have led to the subsequent increase and range expansion of the ticks they host, including A. americanum (Paddock and Yabsley 2007, Virgina Department of Game and Inland Fisheries 2007). To investigate the epidemiological role of A. americanum, we first developed a multiplex real-time PCR assay for Ehrlichiae (E. chaffeensis, E. ewingii, and PME) and three species of Rickettsiae (R. amblyommii, R. parkeri, and R. rickettsii) that could be applied to tick samples. We then collected tick samples from across the state, tested a total of 2545 nymph and adult A. americanum, and compared these findings with state human disease statistics.

Materials and Methods

Study areas and sample collection

The ticks collected for this study came from the VDH, the Fairfax County Department of Health (FCDH), the Old Dominion University (ODU) Department of Biology, and the US Army Public Health Command (USAPHC). Ticks collected by the VDH were obtained between June 1 and August 15, 2012, by cloth drags, primarily from sample sites at selected municipal, county, or state park lands within each of five survey regions of Virginia, including the Richmond City Area, an area in Southside Virginia, the Charlottesville Area, the Lynchburg Area, and the Fredericksburg Area (Fig. 2). Tick samples were prepared as single adult samples (n=112) or as pooled nymphs (n=1206 nymphs in 388 pools; Table 1). Small pools (two ticks/pool) were created to permit the estimation of infection rates for agents that were likely to be common (i.e., >50% infection rates). Large pools contained up to eight ticks and served to increase the number of ticks tested. Ticks collected by VDH were processed and tested at the University of Virginia (UVA).

FIG. 2.

Amblyomma americanum collection sites across Virginia. Richmond City area: 1, Henrico County; 2, Goochland Co.; 3, Chesterfield Co. Charlottesville area: Sites 4–6, Albemarle Co. Southside Virginia area: 7, Brunswick Co.; 8, Mecklenberg Co., and Site 28, Fort Pickett, Nottoway County. Lynchburg area: Sites 9–14, Lynchburg City. Fredericksburg area: 15, Fredericksburg; 16, Stafford Co.; 17, Spotsylvania Co.; 18, King George Co.; Site 27, Fort A.P Hill, Caroline Co.; Fairfax County Sites 19–23 and Site 28, Fort Belvoir (Fairfax Co. and adjacent counties/cities). Hampton Roads Area: 24, Portsmouth City; 25, Cheaspeake City.

Table 1.

Tick Demographics

			Nymphs
	Total samples tested	Adults	Pools	Total
Richmond Area	103	40	63	192
Charlottesville Area	103	19	84	261
Southside Virginia	102	41	61	194
Lynchburg Area	100	9	91	284
Fredricksburg Area	92	3	89	275
Hampton Roads Area	57	45	12	12
Fairfax Area	99	49	50	163
US Army Public Health Command	822	370 (351 pools)	471	588
Overall	1276	576	871	1969

Ticks collected by the FCDH were obtained from five sites within Fairfax County (Fig. 2) by means of cloth drags, flags, or CO₂-baited traps. Ticks were initially submitted to the Johns Hopkins Bloomberg School of Public Health (JHSPH) for DNA extraction, and, subsequently, VDH personnel made a random sampling of tick samples collected between June 4 and August 15, 2012, to be sent to UVA for testing. Nymph extracts were pooled (two to nine extracts per pool), while adults were tested as single samples. A total of 49 single adult ticks and 163 nymphs (in 50 pools) were tested (Table 1).

Ticks from ODU were collected by use of drag cloths from March 1 to July 18, 2012, and came from one site in the City of Chesapeake and one in the City of Portsmouth (Fig. 2). Tick specimens were initially extracted at ODU, and included 45 single adult tick extracts and 12 single nymph extracts from 2012 collections (Table 1). ODU also provided tick extracts (24 adults and 18 nymphs) collected from these two sites in 2010 and 2011.

Ticks from the USAPHC were removed from humans at three army installations: Fort Pickett in Southside Virginia, Fort A.P. Hill in the Fredericksburg Area, and Fort Belvoir in Fairfax County (Fig. 2), and submitted to the USAPHC Tick-borne Disease Laboratory, Aberdeen Proving Ground, Maryland for Ehrlichia testing. Ticks from Fort Pickett and Fort A.P. Hill were removed from soldiers training on those installations, whereas ticks from Fort Belvoir were collected from military personnel, civilian employees, and dependents that worked at that installation but lived in the surrounding areas. USAPHC had “superpooled” extracts from the Virginia ticks with tick extracts from installations in other states. USAPHC also maintained extracts of each individual tick so they could be individually retested when a superpool tested positive. Among these pools, a total of 370 adult ticks (351 pools) and 588 nymphs (471 pools) from Virginia, submitted to USAPHC from March 21, 2012 to September 13, 2012, were tested. Because these superpools were not originally tested for SFGR, it was not possible to back trace the original geographic origin of contributing samples whose superpool later tested positive for a SFGR.

DNA extraction

The Qiagen DNeasy Blood and Tissue Kit (Qiagen Inc., Valencia, CA) was used for ticks processed at UVA or at ODU, with some modifications to the manufacturer's protocol. Briefly, adult ticks (extracted individually) were bisected longitudinally with a razor or surgical blade. Nymphs were extracted individually or in pools. Ticks were placed into 180μL of Buffer ALT and homogenized with glass beads. Samples were centrifuged at 18,000 relative centrifugal force (rcf) for 30 s at room temperature and placed at 56°C in a dry block. For some samples, an additional 180 μL of Buffer ALT was added followed by the addition of 40 μL of Proteinase K Solution and 6 μL of carrier RNA (1 μg/μL; Qiagen). Samples were placed on the side in a shaker incubator at 56°C for a 1- to 3-h digestion. DNA was isolated using the spin-column protocol for animal tissues, eluted, and stored at −20°C prior to testing by PCR.

At JHSPH, DNA was extracted from individual ticks using a MasterPure DNA Purification Kit (Epicentre Biotechnologies, Madison, WI), with a modified procedure. Briefly, a tick was placed in Tissue and Cell Lysis Solution, disrupted with a 5-mm stainless steel bead using a TissueLyser II (Qiagen) for 3 min, then centrifuged for 1 min at 16,100 rcf at room temperature in a microcentrifuge. To each sample, 250 μL of Tissue and Cell Lysis Solution containing 1 μL of Proteinase K (50 g/μL) was added. Beads were removed magnetically, and each sample was incubated at 65°C for 60 min followed by a 5-min incubation on ice. DNA was isolated from debris by addition of 150 μL of MPC Protein Precipitation Reagent and centrifugation at 4°C for 10 min at 17,000 rcf. The DNA was then pelleted using isopropanol, air dried, and then resuspended in 30 μL of molecular-grade water.

At USAPHC, genomic tick DNA was extracted by using a Zymo Genomic DNA II Kit™ (Zymo Research Corporation, Orange, California) according to manufacturers' instructions (Stromdahl et al. 2011).

Detection of bacterial DNA by real-time PCR

All multiplex testing for this study was performed at UVA using the Applied Biosystems ViiA 7 Real-Time PCR system (Applied Biosystems, Foster City, CA). Reactions were performed with Bio-Rad iQ Multiplex Powermix (Bio-Rad, Hercules, CA) in a 25-μL reaction volume with 5 μL of DNA extract. Primers and TaqMan probes (Eurofins MWG Operon, Huntsville, AL) were used at concentrations listed in Table 2. Probes for R. rickettsii and R. amblyommii were adapted from molecular beacon designs. All reactions were run using the following program: 95°C for 3 min, followed by 40 cycles of 95°C for 15 s, and 61°C for 1 min. Each run included a “No template control” (nuclease-free water) and positive controls for each of the six bacterial targets. Postreaction baselines and thresholds were adjusted manually as needed.

Table 2.

Primers and Probes

Bacterial target	Gene target	Oligo name, type	Sequence, 5′→3′	Final conc., μM	Amplicon size	Reference
R. rickettsii	OmpB	RR1370F, F primer	ATAACCCAAGACTCAAACTTTGGTA	0.4	124 bp	Smith et al. 2010
		RR1494R, R primer	GCAGTGTTACCGGGATTGCT	0.4
		RR1425B, Probe	FAM-TTAAAGTTCCTAATGCTATAACCCTTACC-BHQ1	0.2
R. amblyommii	OmpB	Ra477F, F primer	GGTGCTGCGGCTTCTACATTAG	0.2	141 bp	Smith et al. 2010
		Ra618R, R primer	CTGAAACTTGAATAAATCCATTAGTAACAT	0.2
		Ra532P, Probe	HEX-TCCTCTTACACTTGGACAGAATGCT-BHQ2	0.2
R. parkeri	OmpB	Rpa129F, F primer	CAAATGTTGCAGTTCCTCTAAA	0.2	96 bp	Jiang et al. 2010
		Rpa224R, R primer	AAAACAAACCGTTAAAACTACCG	0.2
		RpaTM-P-TAMRA, Probe	TAMRA-AATTAATACCCTTATGARCASCAGCAG-BHQ2	0.2		This study
E. chaffeensis/E. ewingii	16S rRNA	ECH16S-17, F primer	GCGGCAAGCCTAACACATG	0.4	81 bp	Loftis et al. 2003
		ECH16S-97, R primer	CCCGTCTGCCACTAACAATTATT	0.4
E. chaffeensis	16S rRNA	ECH16S-38, Probe	Texas Red-AGTCGAACGGACAATTGCTTATAACCTTTTGGT-BHQ2	0.2
E. ewingii	16S rRNA	EEW16S-97, R primer	CCCGTCTGCCACTAACAACTATC	0.2	83 bp	This study
		EEW16S-P, Probe	Cy5.5-AGTCGAACGAACAATTCCTAAATAGTCTCTGAC-BHQ2	0.4		This study
Panola Mountain Ehrlichia	gltA	Ehr3CS-214F, F primer	TGTCATTTCCACAGCATTCTCATC	0.4	145 bp	Loftis et al. 2008
		PME344R, R primer	ATTAGCGCAATCATACTTGCAA	0.4		This study
		PME266P, Probe	Cy5-TGCCTTAGCTGCACATTATTGTGAT-BHQ2	0.2		This study

Sequencing of R. parkeri samples

Samples positive for R. parkeri in the multiplex assay were retested by singleplex PCR containing the same R. parkeri primers and probe used in the multiplex. Amplicons from the singleplex were cloned into the PCR 2.1 TOPO vector using the TOPO TA cloning kit (Invitrogen, Carlsbad, CA), and resulting plasmids were sequenced using GeneWiz (South Plainfield, NJ) or Eurofins MWG Operon DNA sequencing services. Sequence data were compared to the corresponding ompB gene sequence of R. parkeri strain At24 (Genbank EF102239).

Statistics and calculations

Minimum infection rate (MIR) for pooled adults or nymphs is the number of positive pools per total number of pooled nymphs and assumes that each positive pool of nymphs contains only one infected tick. Maximum likelihood estimation (MLE) was calculated using a the PooledInfRate, ver. 4.0 add-on to Microsoft Excel (Biggerstaff 2009). MLE calculations are based on the number of pools, pool sizes (number of ticks per pool), and number of positive pools. This calculation provides a 95% confidence interval (CI) on the estimate (Biggerstaff 2008). Adult detection rates by region were compared using the Fisher exact test. All p values were two-tailed, and p<0.05 was considered significant.

Results

Test results from ticks collected from the environment by use of dragging or trapping are presented in Tables 3A–C. Test results for ticks collected from personnel on Army Installations are presented in Tables 4A and B.

Table 3A.

Virginia Amblyomma americanum Adults by Survey Regions, Collection Sites and Dates, and Test Results

Tick collection regions, sites and site numbers	Collection dates	Ticks tested	Ehrlichia chaffeensis % Positive	Ehrlichia ewingii% Positive	Panola Mountain Ehrlichia% Positive	Rickettsia amblyommii % Positive	Rickettsia parkeri % Positive
Richmond City		40	5.0	7.5	2.5	72.5	0
Henrico Co. (1)	7/4–6/2012	33	3.0	9.1	3.0	78.8	0
Goochland Co. (2)	7/11–23/2012	6	16.7	0	0	50.0	0
Chesterfield Co. (3)	7/23/12	1	0	0	0	0	0
Charlottesville		19	0	5.3	0	68.4	0
Albemarle Co. (4)	6/10 to 7/20/2012	6	0	0	0	33.3	0
Albemarle Co. (5)	7/19/2012	3	0	0	0	100	0
Albemarle Co. (6)	7/19/2012	10	0	10.0	0	80.0	0
Southside Virginia		41	0	9.8	0	75.6	0
Brunswick Co. (7)	7/25/2012	38	0	10.5	0	79.0	0
Mecklenberg Co. (8)	7/25/2012	3	0	0	0	33.3	0
Lynchburg		9	0	0	0	55.6	0
Lynchburg (9)	7/31/2012	0	—	—	—	—	—
Lynchburg (10)	7/31/2012	3	0	0	0	100	0
Lynchburg (11)	7/31/2012	1	0	0	0	0	0
Lynchburg (12)	7/31/2012	0	—	—	—	—	—
Lynchburg (13)	8/3/2012	3	0	0	0	66.7	0
Lynchburg (14)	8/3/2012	2	0	0	0	0	0
Fredericksburg		3	0	0	0	100	0
Fredericksburg (15)	8/7/2012	1	0	0	0	100	0
Stafford Co. (16)	8/7/2012	2	0	0	0	100	0
Spotsylvania Co. (17)	8/15/2012	0	—	—	—	—	—
King George Co. (18)	8/15/2012	0	—	—	—	—	—
Fairfax County		49	24.5*	14.3	6.1	81.6	2.04
Fairfax Co. (19)	6/15–28/2012	3	33.3	0	0	100	0
Fairfax Co. (20)	6/4 to 7/25/2012	2	100	0	50.0	100	0
Fairfax Co. (21)	6/7 to 7/11/2012	19	26.3	26.3	0	57.9	0
Fairfax Co. (22)	6/4 to 7/25/2012	11	0	18.2	9.1	100	9.1
Fairfax Co. (23)	6/4 to 7/30/2012	14	28.6	0	7.1	92.9	0
Hampton Roads		45	2.2	2.2	0	64.4	2.2
Portsmouth (24)	6/1 to 7/18/2012	28	3.6	3.60	0	71.4	0
Chesapeake (25)	3/23 to 7/18/2012	17	0	0	0	52.9	5.9
Total Across All Sites		206	7.3	7.8	1.9	72.8	1.0

^*p<0.05, for detection rate of E. chaffeensis in Fairfax County area vs. Richmond, Charlottesville, Southside, and Hampton Roads areas.

Table 3B.

Virginia Amblyomma americanum Nymphs by Survey Regions, Collection Sites, Collection Dates, and Test Results for Ehrlichia Species

			Ehrlichia chaffeensis			Ehrlichia ewingii			Panola Mountain Ehrlichia
			Pooled nymphs			Pooled nymphs			Pooled nymphs
				MLEb			MLEb			MLEb
Tick collection regions, sites, and site numbers	Collection dates	Nymphs tested	MIRa% pos.	% pos.	(95% CI)	MIRa% pos.	% pos.	(95% CI)	MIRa% pos.	% pos.	(95% CI)
Richmond City		192	3.7	3.8	(1.7–7.4)	0.5	0.5	(0.0–2.5)	2.1	2.2	(0.7–5.1
Henrico Co. (1)	7/4–6/2012	82	1.2	1.2	(0.1–5.6)	1.2	1.2	(0.1–6.0)	1.2	1.2	(0.1–6.0)
Goochland Co. (2)	7/11–23/2012	54	1.9	1.8	(0.1–8.4)	0	0	—	1.9	1.8	(0.1–8.4)
Chesterfield Co. (3)	7/23/12	56	8.9	10.5	(4.0–22.3)	0	0	—	3.6	3.7	(0.7–11.9)
Charlottesville		261	0	0	—	0.4	0.4	(0.0–1.8)	0.8	0.8	(0.1–2.5)
Albemarle Co. (4)	6/10 to 7/20/2012	14	0	0	—	0	0	—	0	0	—
Albemarle Co. (5)	7/19/2012	101	0	0	—	0	0	—	2.0	2.0	(0.4–6.2)
Albemarle Co. (6)	7/19/2012	146	0	0	—	0.7	0.7	(0.0–3.2)	0	0	—
Southside Virginia		194	0	0	—	0.5	0.5	(0.0–2.4)	1.0	1.0	(0.2–3.4)
Brunswick Co. (7)	7/25/2012	116	0	0	—	0.9	0.9	(0.0–4.0)	0.9	0.9	(0.1–4.2)
Mecklenberg Co. (8)	7/25/2012	78	0	0	—	0	0	—	1.3	1.3	(0.1–5.9)
Lynchburg		284	4.9	5.3	(3.1–8.5)	4.6	5.1	(2.9–8.4)	2.8	3.0	(1.4–5.6)
Lynchburg (9)	7/31/2012	14	7.1	8.3	(0.5–42.6)	0	0	—	0	0	—
Lynchburg (10)	7/31/2012	82	0	0	—	0	0	—	2.4	2.6	(0.5–8.6)
Lynchburg (11)	7/31/2012	12	0	0	—	0	0	—	25.0	27.8	(8.1–60.6)
Lynchburg (12)	7/31/2012	80	1.0	12.0	(5.8–21.9)	11.3	15.7	(7.7–28.7)	2.5	2.6	(0.5–8.2)
Lynchburg (13)	8/3/2012	48	8.3	8.3	(3.0–18.2)	2.1	2.2	(0.1–10.4)	2.1	2.2	(0.1–10.4)
Lynchburg (14)	8/3/2012	48	2.1	2.0	(0.1–9.3)	6.3	6.7	(1.8–17.2)	0	0	—
Fredericksburg		275	0.7	0.7	(0.1–2.4)	0.7	0.7	(0.1–2.3)	1.5	1.5	(0.5–3.5)
Fredericksburg (15)	8/7/2012	14	0	0	—	0	0	—	0	0	—
Stafford Co. (16)	8/7/2012	20	0	0	—	0	0	—	0	0	—
Spotsylvania Co. (17)	8/15/2012	96	2.1	2.1	(0.4–6.9)	0	0	—	0	0	—
King George Co. (18)	8/15/2012	145	0	0	—	1.4	1.4	(0.3–4.4)	2.7	2.9	(0.9–6.8)
Fairfax County		163	11.6	13.4	(8.7–19.6)*	4.3	4.5	(2.0–8.7)	1.2	1.3	(0.2–4.0)
Fairfax Co. (19)	6/15–28/2012	22	4.5	5.1	(0.3–25.0)	9.1	8.8	(1.8–25.3)	0	0	—
Fairfax Co. (20)	6/4 to 7/25/2012	28	10.7	11.4	(3.3–28.2)	0	0	—	3.6	3.8	(0.2–18.4)
Fairfax Co. (21)	6/7 to 7/11/2012	41	7.3	7.2	(2.1–17.8)	2.4	2.5	(0.1–10.7)	2.4	2.3	(0.1–10.7)
Fairfax Co. (22)	6/4 to 7/25/2012	38	13.2	14.0	(5.8–27.8)	5.3	5.2	(1.0–15.7)	0	0	—
Fairfax Co. (23)	6/4 to 7/30/2012	34	20.6	28.6	(14.2–51.6)	5.9	6.4	(1.2–20.3)	0	0	—
Hampton Roads		12	8.3	—c	—	0	0	—	0	0	—
Portsmouth (24)	6/11/2012	4	25.0	—c	—	0	0	—	0	0	—
Chesapeake (25)	6/4 to 10/15/2012	8	0	—c	—	0	0	—	0	0	–
Total across all sites		1381	3.1	3.4e	(2.4–4.3)	1.8	1.9	(1.4–2.7)	1.6	1.6	(1.1–2.4)

^*MLE estimates for Fairfax County area higher than other regions.

^aMIR is minimum infection rate for pooled nymphs; number of positive pools per total number of pooled nymphs; assumes that each positive pool of nymphs contains only one infected nymph.

^bMLE is maximum likelihood estimation from Pooled Infection Rate calculator, Excel add-in; estimates the maximum likely infection rate (%) based on the number of pools, number of ticks per pool, and number of positive pools; provides a 95% confidence interval (CI) on the estimate (Biggerstaff 2009).

^cCalculation of MLE was not necessary because each tested sample consisted of a single tick.

^dAll pools in this group tested positive, making it impossible to estimate a MLE for the testing outcome.

^eThe calculation for MLE-% positive across sites includes the samples from all sites.

Table 3C.

Virginia Amblyomma americanum Nymphs by Survey Regions, Collection Sites, Collection Dates, and Test Results for Rickettsia Species

			Rickettsia amblyommii			Rickettsia parkeri
			Pooled nymphs			Pooled nymphs
				MLEb			MLEb
Tick collection regions, sites, and site numbers	Collection dates	Nymphs tested	MIRa% Pos.	% Pos.	(95% CI)	MIRa% Pos.	% Pos.	(95% CI)
Richmond City		192	30.2	68.3	(55.4–80.9)	0	0	—
Henrico Co. (1)	7/4–6/2012	82	29.3	67.4	(47.6–88.6)	0	0	—
Goochland Co. (2)	7/11–23/2012	54	29.6	61.5	(39.6–84.3)	0	0	—
Chesterfield Co. (3)	7/23/12	56	32.1	72.1	(48.7–94.0)	0	0	—
Charlottesville		261	26.8	54.4	(44.4–65.1)	0	0	—
Albemarle Co. (4)	6/10 to 7/20/2012	14	28.6	44.8	(16.4–87.8)	0	0	—
Albemarle Co. (5)	7/19/2012	101	27.7	56.1	(40.3–73.1)	0	0	—
Albemarle Co. (6)	7/19/2012	146	26.0	53.0	(39.8–67.4)	0	0	—
Southside Virginia		194	24.7	48.4	(37.3–60.5)	0	0	—
Brunswick Co. (7)	7/25/2012	116	25.0	47.9	(34.2–63.6)	0	0	—
Mecklenberg Co. (8)	7/25/2012	78	24.4	47.9	(31.2–67.6)	0	0	—
Lynchburg		284	24.7	46.8	(37.9–56.6)	0	0	—
Lynchburg (9)	7/31/2012	14	21.4	35.8	(10.4–85.6)	0	0	—
Lynchburg (10)	7/31/2012	82	23.2	41.9	(27.2–60.1)	0	0	—
Lynchburg (11)	7/31/2012	12	33.3	39.9	(14.8–74.3)	0	0	—
Lynchburg (12)	7/31/2012	80	25.0	49.1	(32.5–68.5)	0	0	—
Lynchburg (13)	8/3/2012	48	25.0	48.4	(28.0–73.4)	0	0	—
Lynchburg (14)	8/3/2012	48	25.0	48.4	(28.0–73.4)	0	0	—
Fredericksburg		275	27.6	57.5	(47.4–67.9)	0	0	—
Fredericksburg (15)	8/7/2012	14	28.8	44.8	(16.4–87.7)	0	0	—
Stafford Co. (16)	8/7/2012	20	20.0	28.2	(9.9–60.8)	0	0	—
Spotsylvania Co. (17)	8/15/2012	96	26.0	50.4	(35.1–67.7)	0	0	—
King George Co. (18)	8/15/2012	145	29.7	70.9	(55.5–85.6)	0	0	—
Fairfax County		163	28.0	64.4	(49.8–79.8)	3.7	3.8	(1.6–7.7)
Fairfax Co. (19)	6/15–28/2012	22	31.8	—d	—	0	0	—
Fairfax Co. (20)	6/4 to 7/25/2012	28	32.1	—d	—	3.6	3.4	(0.2–15.5)
Fairfax Co. (21)	6/7 to 7/11/2012	41	24.4	49.9	(26.7–78.8)	4.9	5.0	(1.0–15.6)
Fairfax Co. (22)	6/4 to 7/25/2012	38	26.3	49.0	(26.8–78.6)	7.9	8.1	(2.3–20.3)
Fairfax Co. (23)	6/4 to 7/30/2012	34	29.4	—d	—	0	0	—
Hampton Roads		12	50.0	—c	—	0	0	—
Portsmouth (24)	6/11/2012	4	75.0	—c	—	0	0	—
Chesapeake (25)	6/4 to10/15/2012	8	37.5	—c	—	0	0	—
Total across all sites		1381	27.3	55.9e	(51.4–60.5)	0.4	0.4	(0.2–0.9)

^aMIR is minimum infection rate for pooled nymphs; number of positive pools per total number of pooled nymphs; assumes that each positive pool of nymphs contains only one infected nymph.

^bMLE is maximum likelihood estimation from Pooled Infection Rate calculator, Excel add-in; Estimates the maximum likely infection rate (%) based on the number of pools, number of ticks per pool, and number of positive pools; provides a 95% confidence interval (CI) on the estimate (Biggerstaff 2009).

^cCalculation of MLE was not necessary because each tested sample consisted of a single tick.

^dAll pools in this group tested positive, making it impossible to estimate a MLE for the testing outcome.

^eThe calculation for MLE-% positive across sites includes the samples from all sites.

Table 4A.

Amblyomma americanum Adults Submitted to the US Army Public Health Command^a

Submitting installation	Dates received	Adults tested	Ehrlichia chaffeensis Adult ticks% Positiveb	Ehrlichia ewingii Adult ticks% Positiveb	Panola Mountain EhrlichiaAdult ticks% Positiveb
Ft. A. P. Hill	4/3–7/24, 2012	84	1.2	7.1	0
Ft. Belvoir	3/26–8/21, 2012	226	0.4	0.9	0
Ft. Pickett	5/18–8/21, 2012	60	6.7	3.3	0
Total across sites		370	2.8	3.8	0

^aData from USAPHC samples is presented separately due to differences in collection methods.

^bAlthough some of the adult ticks from each installation pooled, none of the pooled adult samples tested positive, permitting the calculation of infection rates as a percentage of the total number of adult ticks tested.

Table 4B.

Amblyomma americanum Nymphs Submitted to US Army Public Health Command^a

			Ehrlichia chaffeensis			Ehrlichia ewingii			Panola Mountain Ehrlichia
			Pooled nymphs			Pooled nymphs			Pooled nymphs
				MLE			MLE			MLE
Submitting installation	Dates received	Nymphs tested	MIR % Pos.	% Pos.	(95% CI)	MIR % Pos.	% Pos.	(95% CI)	MIR % Pos.	% Pos.	(95% CI)
Ft. A. P. Hill	3/21–7/24, 2012	127	0.8	0.8	(0.1–3.8)	0			0.8	0.8	(0.1–3.8)
Ft. Belvoir	3/27–9/13, 2012	194	0.5b			0			0
Ft. Pickett	5/18–8/21, 2012	267	1.5	1.5	(0.5–3.6)	1.5	1.5	(0.5–3.5)	1.1	1.1	(0.3–3.0)
Total across sites		588	0.9	1.0c	(0.4–2.1)	0.5	0.7c	(0.2–1.6)	0.6	0.7c	(0.2–1.6)

^aData from the US Army Public Health Command (USAPHC) samples is presented separately due to differences in collection methods.

^bAlthough some of the nymphs from Fort Belvoir pooled, none of the pooled samples tested positive, so the MIR value represents a percentage of the total number of nymphs tested.

^cThe calculation for MLE was % positive across sites includes the samples from all sites.

MIR, minimum infection rate; MLE, maximum likelihood estimation; CI, confidence interval.

Detection of Ehrlichia species

We observed variable E. chaffeensis infection rates in ticks across regions and sites with ranges of 0–24.5% (average 7.3%) in adult ticks and 0–13.4% (average 3.4%) in nymphs. The Charlottesville Area (Table 3A and B) was unique in that E. chaffeensis infection was absent in both nymph and adult ticks at each of the three sites sampled. E. chaffeensis was also absent in the two sites sampled in Southside Virginia, but was detected in a relatively high proportion of ticks from Fort Pickett within that same region. Adult ticks from Fairfax County had a statistically higher rate of E. chaffeensis infection than was seen from any other region (p<0.05, Fisher exact test) (Table 3A and B), but a surprisingly low E. chaffeensis infection rate (0.4% in adult ticks and 0.5% for nymphs) was observed in the ticks collected from Fort Belvoir personnel who resided in Fairfax County and surrounding areas. The geographic distribution of E. ewingii was more uniform, occurring in adult ticks in all regions except for the Lynchburg and Fredericksburg areas, where the numbers of adult ticks tested was relatively low. The E. ewingii detection rate by region ranged from 2.2% to 14.3% for adult ticks (average 7.8%) and 0.4% to 5.1% (average 1.9%) for nymphs (Tables 3A and B). The statewide adult tick infection rate for E. ewingii averaged 7.8%, similar to the 7.3% statewide E. chaffeensis rate. PME was found in adult ticks at sites in only two regions (Fairfax County and the Richmond City Area), but was found in nymphs at all regions except for the Hampton Roads region. Unexpectedly, although rates of PME infection were relatively high in adult and nymph-stage ticks from most sites in Fairfax County, it was not seen in any of the adult or nymph ticks taken from Fort Belvoir personnel who reside in that same region. Compared to E. chaffeensis and E. ewingii, the percentage of PME-positive samples by region was lower, between 2.5% and 6.1% (average 1.9) for adult ticks and 0.8% and 2.8% (average 1.6) for nymphs.

Detection of Rickettsia species

In addition to the three Ehrlichia species, we detected R. amblyommii and R. parkeri (Table 3A and C), but not R. rickettsii. Adult ticks from all regions and all but three sites tested positive for R. amblyommii, and it was found in nymphs from every region and site sampled. R. amblyommii was found in much greater abundance than any bacterial species tested for, with regional detection rates of between 55.6 and 81.6% (average 72.8%) in adult ticks and 46.8 and 68.3% (average 55.9%) in nymphs. Unexpectedly, R. parkeri was found in A. americanum in two regions of Virginia—in the Hampton Roads region and Fairfax County. Additional tick samples from 2010 and 2011 from the Hampton Roads region were tested to determine if the 2012 R. parkeri–positive ticks were a unique finding. R. parkeri was detected in samples from both the previous 2 years, and samples from all 3 years were confirmed for R. parkeri by singleplex PCR and sequence analysis.

Discussion

This bacterial survey in A. americanum ticks represents the first statewide screening of this species in the state of Virginia since the mid-twentieth century (Sonenshine et al. 1966). We developed a multiplex real-time PCR assay for six bacteria of public health relevance and uncovered several findings.

First, the distribution of E. chaffeensis was focal, but where present rates could be relatively high—up to 24.5% in adults and up to 13.4% in nymphs. Across our study, tick infection rates averaged 7.3% in adults and 3.1 % in nymphs. E. ewingii and PME were generally found at lower rates in nymphs but were present at multiple sites in most regions. Therefore, given the preponderance of A. americanum, and its aggressive predilection for biting humans, clinicians in this region should maintain ehrlichiosis high in the differential diagnosis for tick-borne disease. Human ehrlichiosis cases caused by E. chaffeensis have steadily increased in Virginia each year, from 23 cases in 2007 to 130 cases in 2012. This study provides a baseline, whereby future tick surveillance activities could examine whether increases in human illness are paralleled by increases in the tick population and/or its infection rate.

Second, R. amblyommii was found consistently at sites across the state, with regional rates ranging from 56% to 82% of adults and 47% to 68% of nymphs (MLE estimate, Tables 3A and C). In recent years, Virginia has seen substantial increases in RMSF case detection rates. Specifically, in 2010 a total of 145 cases of RMSF were reported, increasing to 231 cases in 2011 and to 461 cases in 2012; these cases are typically reported to health departments by physicians and/or commercial testing laboratories as RMSF, but have been listed under “Spotted Fever Rickettsiosis” in the VDH annual reports since 2011 (Virginia Department of Health 2012). We think it likely that much of this increase in reported RMSF cases is caused by other illnesses, perhaps Ehrlichia, with “false” cross-reacting seropositivity due to the prevalence of R. amblyommii (Apperson et al. 2008, Smith et al. 2010). The rarity of R. rickettsii detection in D. variabilis (Ammerman et al. 2004, Dergousoff et al. 2009, Moncayo et al. 2010, Stromdahl et al. 2011) and its absence in the lone star tick populations we tested further supports this notion and begs the question of what is the RMSF vector? Again, we would advocate for more aggressive consideration of ehrlichiosis in individuals with clinically suggestive tick-borne illness. Of course this distinction does not change treatment (e.g., doxycycline for either), but could greatly affect public health case reporting and more clearly focus the disease prevention message on A. americanum.

We suspect that deer populations are a significant driver of the A. americanum prevalence in Virginia. Virginia deer populations have risen in the last half-century due to several factors, expanding from an estimated 150,000 in the early 1950s, to approximately 422,000 in 1980, to an estimated 945,000 in 2004 (Virgina Department of Game and Inland Fisheries 2007). This expanding population has resulted in some deer becoming adapted to suburban environments, and control efforts in these settings have been largely unsuccessful. The increase in deer has come with an increase in associated ticks, including A. americanum.

A few other findings are worth noting. The discovery of R. parkeri in A. americanum in Virginia was surprising, but strengthens evidence that A. americanum could be a vector. To our knowledge, this is the first study to identify this pathogen in multiple A. americanum from each of several regions of within a state. The two regions where these R. parkeri–positive A. americanum ticks were found are the only regions known in Virginia to have established A. maculatum populations, suggesting that A. americanum and A. maculatum share one or more reservoir hosts in these two areas of Virginia. Additional studies will be required to determine the identity of these hosts, assess the extent of host sharing, and determine the shared reservoir host range.

We expected to detect PME within Virginia, but we were surprised to detect it in so many regions of the state. A study by Yabsley et al. demonstrated that PME had entered the Virginia white-tailed deer population as early as 2002 and showed these deer to be a natural reservoir host for PME (Yabsley et al. 2008). PME is not only a disease concern in humans, but in dogs as well (Qurollo et al. 2013).Further surveillance for illness caused by PME is needed to understand its significance as a pathogen.

In sum, we developed a six-plex real-time PCR assay to interrogate bacterial populations within the lone star tick, a vector of substantial increase and public health significance in the state of Virginia. We used the assay in tick populations from around the state, resulting in a substantial detection of Ehrlichia, (including rarely considered species), massive detection of R. amblylommii, and occasional detection of R. parkeri. We advocate use of such methods in the future to track changes in pathogen prevalence in this important tick.

Acknowledgements

The project described was supported by grant number K25AI067791 to H.G. from the National Institute Of Allergy and Infectious Diseases (NIAID). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIAID or the National Institutes of Health (NIH). This work was also supported by the Henry M. Jackson Foundation to C.W. Support to D.E.N. for this project comes from Fairfax County, Virginia, and support to T.H. comes from NIH grant number T32 AI 007471, the Johns Hopkins Malaria Research Institute, and the A. Ralph and Silvia E. Barr Fellowship.

Author Disclosure Statement

All authors declare no competing financial interests exist that may create a conflict of interest in connection with this article.