Propagation of the Agent of HGE. Blood (EDTA anticoagulated) was obtained from a Nantucket patient with HGE whose infection was confirmed by blood smear and sequencing of a 16S rDNA amplification product (8), and 0.5 ml was inoculated i.p. into a splenectomized CD-1 mouse. Buffy coat preparations (equivalent to 1 ml of whole blood, centrifuged at 2500 x g for 10 min) were inoculated i.p. into two intact C3H/HeJ mice. Plasma (0.5 ml) was inoculated into one C3H/HeJ mouse as well. To determine whether ehrlichiae were present in the peripheral blood, thin blood smears were taken from the tails of rodents, dried, fixed for 2 min in absolute methanol, and stained with 1 ml Giemsa stock (Harleco Original Azure Blend; EM Diagnostic Systems, Princeton, NJ) diluted in 40 ml distilled water with 150 ptl of 0.5% Na2CO3 and 1.25 ml methanol for 1 hr or more (9

The bacterial 16S rRNA genes from blood samples of two patients with human ehrlichiosis and from an isolate recovered from one of the patients were amplified by using the polymerase chain reaction. The amplimers were then cloned and sequenced. The 16S rRNA gene sequence was also determined for Ehrlichia canis (two strains), E. equi, E. phagocytophila (two strains), and E. sennetsu (two strains). These sequences, along with a previously published 16S rRNA gene sequence of E. risticii, were compared. The 16S rRNA gene sequences were identical for all three sources of the human ehrlichiosis agent. The sequence comparisons indicate that the human ehrlichiosis agent is a new species most closely related to E. canis (98.2%) and more distantly related to other Ehrlichia spp. We propose that this species be named Ehrlichia chaffeensis sp. nov., with the Arkansas strain as the type strain.

Ehrlichia chaffeensis Anderson, Dawson & Wilson, causative agent of human (predominantly monocytic) ehrlichiosis, was successfully transmitted experimentally by Amblyomma americanum (L.) to white-tailed deer, Odocoileus virginianus (Zimmerman). Deer were needle-exposed intravenously to E. chaffeensis in tissue-culture canine macrophage (DH82) cells, and 11 d later were exposed to laboratory-reared A. americanum larvae, nymphs, and adults for acquisition feeding. Three months after this feeding, naive deer and dogs were exposed to recently molted nymphs and adults. Attempted reisolation of the pathogen by way of tissue culture was successful from one needle-exposed deer but not from the tick-exposed deer or dogs. Based on serologic evidence and polymerase chain reaction data, both nymphal and adult ticks transmitted E. chaffeensis to naive deer but not to dogs.

A new disease was recognized in the United States in 1986. The etiologic agent, although not previously isolated from a human, appeared to be serologically related to Ehrlichia canis, a canine leukotropic rickettsia. We obtained blood specimens from 27 febrile patients with a history of tick exposure. Leukocytes from 24 patients not treated with tetracycline were placed onto a monolayer of DH82 cells. We performed indirect immunofluorescence on sera from all 27 febrile patients as well as sera from 12 patients with previously diagnosed ehrlichiosis. Intracytoplasmic inclusions were first observed in culture 35 days after the addition of infected blood from one patient. Partial sequencing of the rRNAs from the human isolate and E. canis indicated that they are 98.7% related. Positive indirect immunofluorescence reactions to the human isolate were obtained for all 12 previously diagnosed patients and for 33% of the 27 febrile patients. Two patients were seropositive for the human isolate but not for E. canis. No sera were positive for E. canis and negative for the human isolate. We report the isolation of a previously unrecognized Ehrlichia sp. that appears to be the etiologic agent of human ehrlichiosis. Serologic data (range of antibody titers, 256 to 32,768) in combination with rRNA sequencing indicated that the newly isolated Ehrlichia sp. is similar, but not identical, to E. canis.

Deer tick-transmitted pathogens such as Lyme disease spirochetes and babesiae appear to require a period of reactivation and replication during the tick's blood meal before it is able to infect a host. The duration of nymphal tick attachment that is required for transmission of the agent of human granulocytic ehrlichiosis (HGE) was determined by removing feeding ticks from mice at various time points. As with spirochetes and babesiae, ehrlichiae infected few mice when ticks were removed prior to 36 h of tick attachment. This "grace period" may serve as a modifying factor in the epidemiology of this newly emergent zoonosis and help physicians make informed decisions concerning management of tick bites in HGE-endemic areas.

The findings of this study are primarily representative of more seriously ill patients with human ehrlichiosis. Although rates are low, ehrlichiosis is found in many areas of the United States. Patients with a history of tick exposure, acute febrile illness, decreasing leukocyte counts, and decreasing platelet counts may have ehrlichiosis. Prompt treatment with tetracycline or chloramphenicol markedly decreases the morbidity.

The roles of wild mammals and ticks in the epidemiology of Ehrlichia chaffeensis at a suspected endemic site were investigated using serologic testing, culture, and polymerase chain reaction (PCR) supported by restriction endonuclease analysis and DNA sequencing. Antibodies reactive to E. chaffeensis (> or = 1:64) were detected in 92% of white-tailed deer (Odocoileus virginianus), 21% of raccoons (Procyon lotor), and 8% of opossums (Didelphis virginianus), but not in 8 other species of mammals. Of 7 species of ticks found by host and environmental sampling, Amblyomma americanum was the dominant species, accounting for greater than 99% of all ticks collected. Deer, raccoons, and opossums were the only species parasitized by all life stages of A. americanum, and A. americanum was the only tick parasitizing deer. A nested PCR protocol incorporating E. chaffeensis-specific primers detected E. chaffeensis DNA in blood, lymph nodes, or spleen from 54% of deer examined. The nested PCR detected E. chaffeensis DNA in 6 of 50 (12%) individual adult A. americanum collected from the environment, in 14 of 79 (18%) pools representing 402 adult A. americanum collected from the environment, and in 7 of 25 (28%) pools of mixed stages of A. americanum collected from deer. Although no Ehrlichia spp. were isolated in culture, sequencing of representative amplicons from deer and ticks confirmed PCR products as E. chaffeensis. These data provide strong evidence that white-tailed deer and lone star ticks are the primary reservoir and vector of E. chaffeensis, respectively. The same PCR protocol, incorporating primers specific for an Ehrlichia-like organism of white-tailed deer, detected this organism in blood, lymph nodes, or spleen from 96% of these deer. The Ehrlichia-like organism of deer was detected by PCR from 0 of 50 individual ticks, 7 of 79 (9%) pools, and 1 of 25 (4%) pools of A. americanum collected from deer. Sequencing of representative amplicons from deer and ticks confirmed PCR products as Ehrlichia-like organism of deer. These data suggest that the Ehrlichia-like organism of deer is present in both the deer and lone star ticks populations at this location.

Although more than 320 cases of human ehrlichiosis have been diagnosed in 27 states since 1986, the reservoir host or hosts remain unknown. Since antibodies reactive to Ehrlichia chajyeensis, the etiologic agent of human ehrlichiosis, have been found in white-tailed deer (Odocoileus virginianus), we experimentally evaluated the susceptibilities of four white-tailed deer to infection with E. chaj'eensis and Ehrlichia canis, a