Abstract

Nymphal Ixodes scapularis ticks were collected from several sites in Rhode Island. Polymerase chain reaction and DNA sequencing were used to determine the presence and prevalence of Anaplasma phagocytophilum human agent (AP-ha) and a genetic variant not associated with human disease (AP-variant 1). The remaining ticks from each cohort were allowed to feed to repletion on either white-footed (Peromyscus leucopus) or DBA/2 (Mus musculus) mice. The engorged ticks and murine blood samples were evaluated for the presence of AP-ha and AP-variant 1. Although a high percentage of the infecting ticks harbored AP-variant 1, only AP-ha was amplified from the murine blood samples. Additional ticks were fed on immunocompromised SCID mice, and, again, only AP-ha was capable of establishing an infection, and only AP-ha could be detected by xenodiagnosis. These data suggest that AP-variant 1 cannot establish an infection in mice, and we propose that AP-variant 1 has an alternative natural reservoir, possibly white-tailed deer

Anaplasma phagocytophilum is an obligate intracellular bacterium that exhibits a tropism for granulocytic leukocytes. Human infections have been referred to as human granulocytic ehrlichiosis because of the previous classification of the agent in the genus Ehrlichia [1]. Human ehrlichiosis presents as an acute febrile illness and can be difficult to diagnose because of the lack of specificity of the clinical signs and symptoms. In the United States, the large majority of infections have occurred in 2 regions, the Upper Midwest and the Northeast, where the agent is transmitted to humans by the tick vector Ixodes scapularis [2–4]

Previous studies have shown that the white-footed mouse (Peromyscus leucopus) is a competent reservoir for A. phagocytophilum [5–7]. However, on the basis of serological assays or the molecular detection of A. phagocytophilum DNA in blood or tissue samples, other small mammals (e.g., squirrels, voles, rats, and raccoons) and large mammals (e.g., white-tailed deer and horses) have also been suggested to be potential reservoirs [7–11]. Most of these molecular-based studies have been based on polymerase chain reaction (PCR) amplification of the 16S rRNA gene of the agent. DNA sequencing of the A. phagocytophilum 16S rRNA gene amplified from ticks in Connecticut and Rhode Island and from white-tailed deer (Odocoileus virginianus) in Maryland and Wisconsin detected a novel genetic variant of A. phagocytophilum AP-variant 1 [8, 12, 13]. The 16S rRNA sequence of AP-variant 1 differs from that originally described by Chen et al. [14], by 2 bp, and the AP-variant 1 sequence has never been amplified from a confirmed human infection. This suggests that AP-variant 1 does not cause human infections and that this strain may show additional biological or genetic properties distinct from those of the human disease–causing agent. AP-variant 1 was found to be common in ticks in Rhode Island, where few cases of human granulocytic ehrlichiosis have been reported, but rare in Connecticut, where more cases have been reported than in any other state [12]. The prevalence of AP-variant 1 in Rhode Island may be due to a selective advantage over the human agent (AP-ha), in either vector or reservoir populations. Despite the analysis of a large number of blood samples from white-footed mice from both Rhode Island and Connecticut, AP-variant 1 has not been found in white-footed mice [12, 15]. The present study was designed to test the infectivity of AP-variant 1 for white-footed mice and for other commonly used strains of laboratory mice. If infections can be established, these mice would then provide blood and tissue samples for subsequent attempts to obtain an isolate of the variant in tissue culture

Materials and Methods

Mouse strainsAnimals used for experimental procedures included the white-footed mouse and 2 in-bred strains of laboratory mice: immunocompetent DBA/2 and immunocompromised CB-17/SCID

Tick collection, prevalence testing, and mouse infestation Questing nymphal black-legged ticks (I. scapularis) were collected from 4 sites (Corrano, Tower, Trustom, and Webster) near South Kingston, Rhode Island, in June 2001 by following standardized sampling procedures [16]. To determine the prevalence of AP-ha and AP-variant 1, ∼30 ticks from each site were immediately used for DNA extraction, PCR testing, and DNA sequencing. Additional ticks that were collected from each site were maintained in a humidified chamber. The prevalence of AP-ha and AP-variant 1 in the flat ticks allowed us to calculate the number of ticks necessary to ensure ⩽1 infected tick/mouse. This number of ticks (6–8 Trustom ticks and 10–12 Corrano ticks) were placed on each mouse and were allowed to feed to repletion. The engorged nymphs were frozen immediately and were stored at −20°C until used for DNA extraction

Xenodiagnosis and collection of murine blood samples Uninfected, laboratory-reared I. scapularis larvae were placed on mice on days 10 and 17 after mice had been subjected to infestation with field-collected ticks; ticks were allowed to feed to repletion. Engorged larvae were collected and were stored in a humidified chamber until they molted to the nymphal stage. DNA was extracted from these newly molted nymphs and was used for PCR testing, and the amplicons from positive samples were used for DNA sequencing. Blood samples were collected from mice by retroorbital bleeding. Blood samples were either used for DNA extractions immediately or stored at −20°C until DNA was extracted

DNA extractionsDNA was extracted directly from blood samples by use of a QIAamp DNA Blood Mini Kit (Qiagen); the protocol followed was suggested by the manufacturer. In brief, detergent lysis was performed in the presence of proteinase K, for 10 min at 70°C. The lysed material was applied to a spin column containing a silica gel–based membrane and was washed twice. Purified DNA was eluted from the columns in sterile distilled H2O and was stored at 4°C until used as a template for PCR amplification. DNA was extracted from I. scapularis ticks by a modification of the manufacturer’s protocol for the DNeasy Tissue Kit (Qiagen), as described elsewhere [13]

PCR analysisA nested PCR assay that amplified a 546-bp portion of the 5′ region of the 16S rRNA gene was used to identify A. phagocytophilum in tick and murine blood samples [13]. In brief, primary amplifications consisted of 40 cycles in an ABI GeneAmp PCR System 9700 thermal cycler (Applied Biosystems), with each cycle consisting of denaturation for 30 s at 94°C, annealing for 30 s at 55°C, and extension for 1 min at 72°C. The 40 cycles were preceded by denaturation for 2 min at 95°C and were followed by extension for 5 min at 72°C. Primary amplifications used primers ge3a and ge10r and reagents from the Qiagen Taq PCR Master Mix Kit (Qiagen). Each reaction contained 2.5 μL of purified DNA as template in a total volume of 25 μL and 200 μmol/L each dNTP (dATP, dCTP, dGTP, and dTTP), 1.25 U of Taq polymerase, and 0.5 μmol/L each primer. Reaction products were subsequently maintained at 4°C until used as a template for nested reactions

Nested amplifications used primers ge9f and ge2 and 1 μL of the primary PCR product as a template, in a total volume of 50 μL. Each nested amplification contained 200 μmol/L each dNTP (dATP, dCTP, dGTP, and dTTP), 1.25 U of Taq polymerase, and 0.2 μmol/L each primer. Nested cycling conditions were as described for the primary amplification, except 30 cycles were used. Reactions were subsequently maintained at 4°C until analyzed by agarose-gel electrophoresis or purified for DNA sequencing

DNA sequencing and data analysisAll samples producing positive PCR products were subjected to DNA sequencing reactions by use of fluorescent-labeled dideoxynucleotide technology (BigDye Terminator Cycle Sequencing Ready Reaction Kit; Applied Biosystems). Sequencing reaction products were separated, and data were collected by use of an ABI 3100 Genetic Analyzer automated DNA sequencer (Applied Biosystems). To ensure maximum accuracy of the data, the full sequence was determined for both strands of each DNA template. Sequences were edited and assembled by the Staden software programs [17] and were analyzed by the Wisconsin Sequence Analysis Package (Genetics Computer Group) [18]. The partial 16S rRNA sequence for AP-variant 1 has been deposited in GenBank (accession no. AY193887)

Results

Determination of prevalence of AP-ha and AP-variant 1 in ticksPCR and DNA sequencing of the 16S rRNA gene was used to determine the prevalence of AP-ha and AP-variant 1 in a subset of cohorts of questing nymphal I. scapularis ticks (n=116) that had been collected from 4 sites in Rhode Island. Each site tested had previously been shown to have high tick and reservoir densities. The results of this preliminary testing are shown in table 1. The Trustom site had both the highest total prevalence of infected ticks (25.8%) and the highest prevalence of AP-variant 1 (22.6%). No variants of A. phagocytophilum other than AP-variant 1 were found in any of the ticks. Because AP-ha was not detected at either the Tower or Webster sites, ticks from these 2 sites were not used for subsequent feedings on mice

Table 1

Prevalence of Anaplasma phagocytophilum human agent (AP-ha) and A. phagocytophilum variant 1 (AP-variant 1), in nymphal Ixodes scapularis ticks collected from 4 sites in Rhode Island

Table 1

Prevalence of Anaplasma phagocytophilum human agent (AP-ha) and A. phagocytophilum variant 1 (AP-variant 1), in nymphal Ixodes scapularis ticks collected from 4 sites in Rhode Island

Tick infestations of miceNymphal ticks collected from the Corrano (10–12 ticks/mouse) and Trustom (6–8 ticks/mouse) sites were placed on A. phagocytophilum–naive laboratory-bred mice (white footed [n=32] or DBA/2 [n=30]), were allowed to feed to repletion, and were collected in water pans. A total of 337 engorged ticks were collected and tested by PCR and DNA sequencing for the presence of AP-ha and AP-variant 1 (tables 2–5). The number of ticks collected from individual mice ranged from 1 to 10. The DNA sequencing results from the engorged ticks identified which agents were presented to each mouse. A total of 32 of the engorged nymphs were positive for AP-variant 1, and 15 were positive for AP-ha. A total of 3 of the AP-ha–infected mice (DBA-7, DBA-8, and DBA-24) had ticks without molecular evidence of AP-ha infection (tables 3 and 4)

Table 2

Results from ticks collected at the Trustom site that fed to repletion on naive white-footed mice and from blood samples from white-footed mice

Table 2

Results from ticks collected at the Trustom site that fed to repletion on naive white-footed mice and from blood samples from white-footed mice

Table 3

Results from ticks collected at the Trustom site that fed to repletion on naive DBA/2 mice and from blood samples from DBA/2 mice

Table 3

Results from ticks collected at the Trustom site that fed to repletion on naive DBA/2 mice and from blood samples from DBA/2 mice

Table 4

Results from ticks collected at the Corrano site that fed to repletion on either naive white-footed or DBA/2 mice and from blood samples from white-footed or DBA/2 mice

Table 4

Results from ticks collected at the Corrano site that fed to repletion on either naive white-footed or DBA/2 mice and from blood samples from white-footed or DBA/2 mice

Table 5

Results from field-collected nymphal ticks that fed on SCID mice, from blood samples from SCID mice, and from xenodiagnostic ticks

Table 5

Results from field-collected nymphal ticks that fed on SCID mice, from blood samples from SCID mice, and from xenodiagnostic ticks

Detection ofA. phagocytophilumin murine blood samples EDTA-anticoagulated whole blood samples were collected from each mouse on days 10 and 17 after tick attachment and were tested for AP-ha and AP-variant 1. The results of this testing are shown in tables 2 (Trustom ticks fed on white-footed mice), 3 (Trustom ticks fed on DBA/2 mice), and 4 (Corrano ticks fed on either white-footed or DBA/2 mice). Although 12 white-footed mice were fed upon by PCR-positive Trustom ticks (table 2), only 1 of these ticks was positive for AP-ha, and none of the mice had positive results for either agent, by PCR testing of blood samples. Five DBA/2 mice were positive by PCR after having been fed upon by Trustom ticks (table 3), and the blood samples from each of these mice contained only AP-ha. Four white-footed and 3 DBA/2 mice were positive by PCR after having been upon by Corran ticks (table 4), and the blood samples from each of these mice contained only AP-ha. In total, 12 mice (4 white footed and 8 DBA/2) became infected, and each of these infections was with AP-ha. Of the 12 mice exposed to AP-ha, on the basis of the engorged-tick results, 9 became infected with AP-ha. In contrast, of the 22 mice exposed to AP-variant 1, none became infected

Infections in SCID miceConsidering that the mouse immune response may be responsible for our inability to detect an infection by AP-variant 1 in immunocompetent white-footed and DBA/2 mice, additional field-collected ticks were placed on 5 SCID mice (8 Trustom ticks/mouse, 3 mice; and 13 Corrano ticks/mouse, 2 mice). As described above, ticks and blood samples from mice were tested for AP-ha and AP-variant 1, by PCR and DNA sequencing (table 5). Although each of the 5 mice was fed upon by at least 1 infected tick, 2 mice were fed upon by ticks that were positive for both AP-ha and AP-variant 1 (SCID-3 and SCID-4), and both of these mice became positive for AP-ha only. These 2 mice had positive results for AP-ha in both blood samples collected (day 5, both mice; day 12, SCID-3; day 14, SCID-4). Although each of the 5 SCID mice was fed upon by a tick positive for AP-variant 1, only 1 mouse (SCID-1) became positive for AP-variant 1, and this positive result by PCR was of the blood sample drawn at the early time point (day 5) only. The sample drawn on day 14 yielded negative results

Xenodiagnosis of SCID miceBecause it appeared that AP-variant 1 may have transiently infected 1 of the 5 SCID mice, xenodiagnosis was performed on each of these mice. Uninfected, laboratory-reared I. scapularis larvae were placed on mice on day 7 after nymphal attachment and were allowed to feed to repletion. Engorged larvae were allowed to molt to nymphs, and DNA was extracted from pools of 5 nymphs and was tested by PCR and DNA sequencing. All 4 pools of ticks that had fed on the mouse positive for AP-variant 1, SCID-1, were negative. In contrast, each of the 8 pools that had fed on the mice positive for AP-ha, SCID-3 and SCID-4, were positive for AP-ha. Tick pools that had fed on the PCR-negative mice, SCID-2 and SCID-5, were negative

Discussion

Several apparent discrepancies between the results for engorged ticks and those for murine blood samples are evident in tables 3 and 4. For example, 3 mice (DBA-7, DBA-8, and DBA-24) became positive for AP-ha, even though each of the engorged nymphs tested were negative for AP-ha. Although a fixed number of ticks were placed on each mouse, only ticks that fed to repletion and dropped off unassisted were collected and tested. Numerous partially fed ticks were not included in the analysis because the mice removed them while grooming themselves. Therefore, the infections in these 3 mice were likely to have been introduced by partially fed ticks that were not collected. Three additional mice (PL-6, PL-23, and DBA-9) were negative for AP-ha when tested on both days 7 and 14 but had been fed upon by at least 1 tick that tested positive for AP-ha. These negative PCR results of murine blood samples may be due to collecting samples outside the window of infection, inefficient transmission of the agent from the tick to the rodent during feeding, or lack of sensitivity of the assay used for analysis. However, the latter seems unlikely, because the PCR assay used for analysis has been shown to detect <2 copies of the 16S rRNA gene [13]

In the attempts to infect SCID mice, 1 mouse was positive by PCR for AP-variant 1 on day 5 after nymphal attachment. This was the only mouse that was positive by PCR for AP-variant 1 throughout the course of the present study. However, this mouse was negative by PCR on day 14, and xenodiagnostic tick testing also yielded negative results. Therefore, it may be that the AP-variant 1 DNA that was amplified on day 5 was derived from the organisms that were injected by a feeding tick and not cleared, probably because of the compromised immune system of this mouse rather than replication of the agent in the mouse. This inference is supported by the negative results by xenodiagnosis. Alternatively, a situation may have occurred wherein AP-variant 1 underwent limited replication in this SCID mouse and the infection was cleared before day 14 and xenodiagnostic testing

Our results demonstrating the inability of ticks harboring AP-variant 1 to transmit the agent to mice can lead to 1 of 2 conclusions: (1) that mice are not susceptible to infection or (2) that AP-variant 1 in the ticks was not viable. However, several factors strongly suggest that the latter was not the case. Quantitative PCR on the flat nymphal ticks indicated that the number of copies of AP-variant 1 DNA in the AP-variant 1–infected ticks was comparable to the amount of AP-ha in AP-ha–infected ticks, >20,000 copies/tick (data not shown). Any agent detected in a flat nymphal tick could come only from the previous life stage (larvae). The nymphs in the present study were collected in early June, which means that they had fed as larvae in the late summer or early fall of the previous year. Therefore, the agent that was detected in the flat nymphs was maintained in these ticks for at least 8 months and through the molt from larval to nymphal stage. Additionally, quantitative PCR showed significantly higher average amounts (>7-fold increase) of AP-variant 1 in engorged ticks than in flat ticks, suggesting replication of the agent induced by tick feeding (data not shown). On the basis of these factors, we conclude that the AP-variant 1 that was present in the ticks was viable and capable of infecting a susceptible host

Because A. phagocytophilum is not transmitted transovarially from 1 tick generation to the next, the maintenance of this species in nature requires competent reservoirs and efficient vectors. I. scapularis ticks have been shown to be an efficient vector [2, 4, 6], and various species of large and small mammals have been examined as potential reservoirs [7–11, 19]. Although these studies have shown that deer, raccoons, squirrels, chipmunks, dogs, and other mammals can be infected with A. phagocytophilum in nature, only the white-footed mouse has been involved in controlled laboratory studies of reservoir competence, and A. phagocytophilum strains used in these studies were either isolated from a human patient or later classified as AP-ha by DNA sequencing [6, 20]. More-recent studies have suggested that a mixed population of A. phagocytophilum strains, with varying host tropisms, may exist in nature [12]. AP-variant 1 has been described in I. scapularis ticks in Connecticut and Rhode Island and in white-tailed deer in Maryland and Wisconsin [8, 12, 13]. AP-variant 1 has not been reported from studies of blood samples from white-footed mice. However, most of these studies were based on serological detection methods that would likely not differentiate AP-ha from AP-variant 1. Our results clearly demonstrate that AP-variant 1 does not infect either white-footed mice or a commonly used laboratory strain, DBA/2. These data suggest that white-footed mice are not a natural reservoir for AP-variant 1 and that other reservoir species are required for maintenance of this strain in the wild. The detection of AP-variant 1 DNA in white-tailed deer suggests that they may serve as a primary reservoir for this agent. In a previous study of 32 blood samples collected from white-tailed deer in Maryland, each of the 3 samples with positive results contained AP-variant 1 [13]. In contrast, the human agent, AP-ha, has never been detected in a white-tailed deer [8, 13]. Maintenance of a natural cycle of infection of AP-variant 1 with deer as the only reservoir is a distinct possibility, because white-tailed deer are a host for all active stages of I. scapularis ticks (larval, nymphal, and adult) [21]

The known cocirculation of the AP-ha and AP-variant 1 strains of A. phagocytophilum in I. scapularis ticks has several implications. Because AP-variant 1 seems to be the predominant strain in areas where both strains are found, it has been previously suggested that the variant may have a competitive advantage in nature, in either the reservoir or vector [12]. Because we have shown that AP-variant 1 will not infect white-footed mice, the presumed primary reservoir of AP-ha, I. scapularis ticks, remain the only known common factor in the natural cycles of both strains. Therefore, if a competitive advantage does exist for AP-variant 1, it would likely occur in the tick. However, testing of this hypothesis in the laboratory will require an isolate of AP-variant 1, which is currently not available. The tick source for infecting laboratory-reared mice was host-seeking I. scapularis nymphs that had fed a single time as larvae, collected from either of 2 sites in Rhode Island. Both AP-ha and AP-variant 1 were present at each of these sites, as evidenced by the detection of both strains in flat ticks. However, no dual infections were detected in any of the flat (n=116) or engorged nymphs (n=337) that were tested. These results suggest several possibilities, one being that the larvae fed on reservoirs harboring only 1 strain. In this scenario, ticks that had fed on either white-footed mice or other reservoirs that harbored only AP-ha would be positive for AP-ha, and ticks that had fed on white-tailed deer or other reservoirs that contained only AP-variant 1 would be positive for the AP-variant 1. Alternatively, larvae may have fed on reservoirs that support both AP-ha and AP-variant 1, but 1 strain precludes growth of the other in ticks. The latter may result from a competitive advantage of 1 strain over the other, as mentioned above, or could be an artifact related to the relative number of organisms of each strain that were ingested by the tick. In the case of a mixed infection, the PCR assay used for detection will favor the predominant agent. For example, when templates are mixed at a 1:9 ratio and amplified in a single reaction, sequencing of the amplicon will not detect the lower-concentration template (data not shown). Because these ticks were field-collected and likely to have fed as larvae on various hosts, a combination of these factors may be involved in our inability to detect dual infections in ticks. However, in a previous study in which 232 adult I. scapularis ticks from the Trustom site in Rhode Island were tested, no dual infections were detected among the 52 ticks with positive results by PCR that were analyzed by DNA sequencing [12]. Because these were adult ticks, they had been fed 2 blood meals and were potentially exposed to infection twice. This suggests that ticks may be resistant to superinfection by >1 A. phagocytophilum strain. There is a precedent for resistance to superinfection among Anaplasma species, because a recent study by de la Fuente et al. [22] has demonstrated that Anaplasma marginale the etiologic agent of bovine anaplasmosis, resists superinfection in both cultured tick cells and bovine erythrocytes. In the case of AP-variant 1, if white-tailed deer are the primary reservoir and infected ticks are resistant to superinfection, larvae feeding on deer positive for AP-variant 1 will acquire the agent and be resistant to AP-ha at subsequent stages. If this is true, an increased prevalence of AP-variant 1 in the deer population, combined with the steadily increasing deer population in many areas of the United States, could affect the distribution and prevalence of AP-ha in nature. However, additional studies are needed to confirm the reservoir competence of white-tailed deer for AP-variant 1, to examine the reservoir potential of other species of small and large mammals, and to understand the interaction of AP-ha and AP-variant 1 in I. scapularis ticks and in both known and potential reservoir populations

Last, the 2-bp difference in the 16S rRNA gene sequences, between AP-ha and AP-variant, 1 serves as a molecular marker that allows us to differentiate these 2 strains but is unlikely to result in any biological significance. The altered reservoir competence of these strains more likely results from changes in genes encoding structural and outer-membrane proteins that may affect the tissue tropism and infectivity of the agent in a particular host. An isolate of AP-variant 1 will be instrumental for facilitating future genetic and biological studies that will allow us to assess the factors that make AP-variant 1– and AP-ha–unique strains of A. phagocytophilum

Acknowledgments

We thank Kim Slater, Danielle Ross, and Khadeja Haye, for excellent technical assistance; Gregory Dasch, for manuscript review; and the Centers for Disease Control and Prevention Biotechnology Core Laboratory, for the synthesis of DNA primers used in this study

References

1
Dumler
JS
Barbet
AF
Bekker
CPJ
, et al.  . 
Reorganization of genera in the families Rickettsiaceae and Anaplasmataceae in the order Rickettsiales: unification of some species of Ehrlichia with Anaplasma, Cowdria with Ehrlichia and Ehrlichia with Neorickettsia descriptions of six new species combinations, and designation of Ehrlichia equi and “HGE agent” as subjective synonyms of Ehrlichia phagocytophilum
Int J Syst Evol Microbiol
 , vol. 
51
 (pg. 
2145
-
65
)
2
Magnarelli
LA
Stafford KC
III
Mather
TN
Yeh
MT
Horn
KD
Hemocytic rickettsia-like organisms in ticks: serologic reactivity with antisera to Ehrlichiae and detection of DNA of agent of human granulocytic ehrlichiosis by PCR
J Clin Microbiol
 , vol. 
33
 (pg. 
2710
-
4
)
3
McQuiston
JH
Paddock
CD
Holman
RC
Childs
JE
The human ehrlichioses in the United States
Emerg Infect Dis
 , vol. 
5
 (pg. 
635
-
42
)
4
Pancholi
P
Kolbert
CP
Mitchell
PD
, et al.  . 
Ixodes dammini as a potential vector of human granulocytic ehrlichiosis
J Infect Dis
 , vol. 
172
 (pg. 
1007
-
12
)
5
Magnarelli
LA
Anderson
JF
Stafford KC
III
Dumler
JS
Antibodies to multiple tick-borne pathogens of babesiosis, ehrlichiosis, and Lyme borreliosis in white-footed mice
J Wildl Dis
 , vol. 
33
 (pg. 
466
-
73
)
6
Telford SR
III
Dawson
JE
Katavolos
P
Warner
CK
Kolbert
CP
Perpetuation of the agent of human granulocytic ehrlichiosis in a deer tick-rodent cycle
Proc Natl Acad Sci USA
 , vol. 
93
 (pg. 
6209
-
14
)
7
Walls
JJ
Greig
B
Neitzel
DF
Dumler
JS
Natural infection of small mammal species in Minnesota with the agent of human granulocytic ehrlichiosis
J Clin Microbiol
 , vol. 
35
 (pg. 
853
-
5
)
8
Belongia
EA
Reed
KD
Mitchell
PD
, et al.  . 
Prevalence of granulocytic Ehrlichia infection among white-tailed deer in Wisconsin
J Clin Microbiol
 , vol. 
35
 (pg. 
1465
-
8
)
9
Levin
ML
Nicholson
WL
Massung
RF
Sumner
JW
Fish
D
Comparison of the reservoir competence of medium-sized mammals and Peromyscus leucopus for Anaplasma phagocytophila in Connecticut
Vector Borne Zoonotic Dis
 , vol. 
2
 (pg. 
125
-
36
)
10
Magnarelli
LA
Ijdo
JW
Stafford KC
III
Fikrig
E
Infections of granulocytic ehrlichiae and Borrelia burgdorferi in white-tailed deer in Connecticut
J Wildl Dis
 , vol. 
35
 (pg. 
266
-
74
)
11
Nicholson
WL
Castro
MB
Kramer
VL
Sumner
JW
Childs
JE
Dusky-footed wood rats (Neotoma fuscipes) as reservoirs of granulocytic ehrlichiae (Rickettsiales: Ehrlichieae) in northern California
J Clin Microbiol
 , vol. 
37
 (pg. 
3323
-
7
)
12
Massung
RF
Mauel
MJ
Owens
JH
, et al.  . 
Genetic variants of Ehrlichia phagocytophila Rhode Island, and Connecticut
Emerg Infect Dis
 , vol. 
8
 (pg. 
467
-
72
)
13
Massung
RF
Slater
K
Owens
JH
, et al.  . 
Nested PCR assay for detection of granulocytic ehrlichiae
J Clin Microbiol
 , vol. 
36
 (pg. 
1090
-
5
)
14
Chen
SM
Dumler
JS
Bakken
JS
Walker
DH
Identification of a granulocytotropic Ehrlichia species as the etiologic agent of human disease
J Clin Microbiol
 , vol. 
32
 (pg. 
589
-
95
)
15
Stafford KC
III
Massung
RF
Magnarelli
LA
Idjo
JW
Anderson
JF
Infection with agents of human granulocytic ehrlichiosis, Lyme disease, and babesiosis in white-footed mice (Peromyscus leucopus) in Connecticut
J Clin Microbiol
 , vol. 
37
 (pg. 
2887
-
92
)
16
Nicholson
MC
Mather
TN
Methods for evaluating Lyme disease risks using geographic information systems and geospatial analysis
J Med Entomol
 , vol. 
33
 (pg. 
711
-
20
)
17
Dear
S
Staden
R
A sequence assembly and editing program for efficient management of large projects
Nucleic Acids Res
 , vol. 
19
 (pg. 
3907
-
11
)
18
Devereux
J
Haeberli
P
Smithies
O
A comprehensive set of sequence analysis programs for the VAX
Nucleic Acids Res
 , vol. 
12
 (pg. 
387
-
95
)
19
Greig
B
Asanovich
KM
Armstrong
PJ
Dumler
JS
Geographic, clinical, serologic, and molecular evidence of granulocytic ehrlichiosis, a likely zoonotic disease, in Minnesota and Wisconsin dogs
J Clin Microbiol
 , vol. 
34
 (pg. 
44
-
8
)
20
Levin
ML
Fish
D
Immunity reduces reservoir host competence of Peromyscus leucopus for Ehrlichia phagocytophila
Infect Immun
 , vol. 
68
 (pg. 
1514
-
21
)
21
Anderson
JF
Mammalian and avian reservoirs for Borrelia burgdorferi
Ann NY Acad Sci
 , vol. 
539
 (pg. 
180
-
91
)
22
de la Fuente
J
Garcia-Garcia
JC
Blouin
EF
Saliki
JT
Kocan
KC
Infection of tick cells and bovine erythrocytes with one genotype of the intracellular ehrlichia Anaplasma marginale excludes infection with other genotypes
Clin Diagn Lab Immunol
 , vol. 
9
 (pg. 
658
-
68
)
Presented in part: International Conference on Rickettsiae and Rickettsial Diseases, Ljubljana, Slovenia, 4–7 September 2002 (abstract O-11)
Financial support: National Institutes of Health (grant AI-30733)