Clinical outcomes of solid organ transplant recipients with ehrlichiosis.

Because of our experience with severe Ehrlichia infections in lung transplant recipients, we reviewed all cases of ehrlichiosis in solid organ transplant recipients at Barnes-Jewish Hospital in St. Louis, Missouri. Between 1996 and 2007, 25 cases of ehrlichiosis were identified. We retrospectively collected demographic, clinical, laboratory, and outcomes data, and we compared the 5 cases in lung transplant recipients with 20 cases in other solid organ transplant recipients (heart, 2; kidney, 13; liver, 5). The presenting symptoms in the majority of both groups consisted of fever and headache. Clinical outcomes were worse in the lung transplant group and included a greater need for intensive care unit treatment (80% vs. 20%, P=0.02), longer length of hospital stay (21 vs. 5 days, P=0.02), and propensity to develop acute lung injury or acute respiratory distress syndrome (60% vs. 10%, P=0.04). No mortalities occurred in either group of patients. In an endemic area, ehrlichiosis is not unusual in solid organ transplant recipients, and lung transplant recipients tend to have a more severe illness.

Application of the polymerase chain reaction for the detection of Ehrlichia canis in tissues of dogs.

Doxycycline treatment resulted in a carrier status in 3 dogs and complete clearance of Ehrlichia canis in 2 dogs (Iqbal and Rikihisa, 1994). Using specimens obtained during that study applicability of polymerase chain reactions (PCRs) in detecting E. canis DNA in tissue specimens and correlation of PCR results with our previous cell culture isolation results were evaluated. PCRs using a pair of primers specific to E. canis 16SrRNA gene sequence were used to detect DNA of E. canis in tissues of 5 experimentally-infected dogs 2 months after doxycycline treatment. An approximately 600 bp product defined by the specific primers was amplified in blood, kidney, lymph nodes, liver, and/or spleen of 3 dogs from which E. canis was reisolated in cell culture. In contrast, E. canis DNA was not detected in tissue or blood specimens of the 2 dogs from which E. canis was not reisolated after doxycycline treatment or in 2 control uninfected dogs. The findings indicate PCR is effective in detecting E. canis in tissues.

Experimental transmission of Ehrlichia canis (Rickettsiales: Ehrlichieae) by Dermacentor variabilis (Acari: Ixodidae).

Four trials were conducted in which laboratory-reared Dermacentor variabilis nymphs were exposed to Ehrlichia canis by feeding on experimentally infected dogs as soon as classical morulae were detected in peripheral blood monocytes. After molting 25, 50 or 90 adult tick pairs were permitted to feed on 7 Ehrlichia-naive dogs. Transmission occurred in trials 1 (1/1 dog), 3 (1/1 dog) and 4 (2/2 dogs) but not in trial 2 (0/3 dogs), with 4 of 7 dogs becoming infected. Successful transstadial transmission was demonstrated by detection of morulae in peripheral blood lymphocytes and by seroconversion to Ehrlichia canis 30 d post-exposure. Incubation periods ranged between 17 and 22 days (mean = 19). Clinical signs, typical of ehrlichiosis, included mucopurulent ocular discharge, lymphadenopathy and malaise with accompanying pyrexia, leukopenia and thrombocytopenia. Pyrexia, thrombocytopenia and erythrophagocytosis and vacuolization of the cytoplasm of monocytic cells were observed 1-4 d prior to detection of morulae. This is the first demonstration that a tick other than Rhipicephalus sanguineus is capable of transstadial transmission of this important pathogen of dogs.

Inhibition of phagosome-lysosome fusion in ovine polymorphonuclear leucocytes by Ehrlichia (Cytoecetes) phagocytophila.

Ehrlichia (Cytoecetes) phagocytophila, the causative agent of tick-borne fever, is an intracellular bacterium that survives and multiplies within granulocytes and monocytes. In the present study, the possible fusion of lysosomes with phagosomes containing E. phagocytophila was investigated in poly-morphonuclear (PMN) cells of sheep infected with the agent, acid phosphatase cytochemistry and cationized ferritin being used as markers of primary and secondary lysosomal enzymes. Latex beads or Candida albicans were incubated with infected and uninfected PMN cells and labelled with the same lysosomal markers. Lysosomal enzymes labelled with the markers were commonly found in phagosomes containing latex beads or C. albicans, but there was no evidence of phagosome-lysosome (P-L) fusion in phagosomes containing E. phagocytophila. It was significant that in cells that contained E. phagocytophila, latex beads and C. albicans, P-L fusion occurred only in phagosomes containing latex beads or C. albicans. However, evidence of P-L fusion with phagosomes containing E. phagocytophila was obtained when PMN cells were incubated with oxytetracycline, which is known to inhibit synthesis of bacterial proteins. These findings indicate that E. phagocytophila is capable of inhibiting P-L fusion and that oxytetracycline depresses this capability.

Two-step PCR in the evaluation of antibiotic treatment for Ehrlichia platys infection.

We evaluated the feasibility of using the two-step polymerase chain reaction (PCR) in determining the withdrawal time of antibiotic treatment for Ehrlichia platys infection. We also present experimental evidence of a dog remaining a carrier after treatment with tetracycline. Canine infectious cyclic thrombocytopenia (CICT) was induced in 3 dogs by intravenous inoculation of blood infected with E. platys. Tetracycline was administered to one of the dogs for 2 weeks when parasitemia appeared. Although the hematologic abnormality of cyclic thrombocytopenia soon disappeared, a few parasitized platelets reappeared after the withdrawal of treatment, and the dog thus remained as a carrier. The other dogs were treated with doxycycline when parasitemic episodes first developed. The durations of antibiotic regimens were determined by the results of two-step PCR in which the 16S rDNA of E. platys was amplified from blood samples. Doxycycline was withdrawn after 8 days of treatment, and the follow-up monitoring continued for 3 weeks. The platelet counts of the 2 dogs remained within the normal range, and the etiologic agent of CICT was not found either by Giemsa staining or by the two-step PCR, indicating complete elimination of the agent.

The effect of two different oxytetracycline treatments in experimental Ehrlichia phagocytophila infected lambs.

The effect of 2 different oxytetracycline treatments in acute E. phagocytophila infected lambs was investigated. Twenty 5-month-old lambs of the Dala and Rygja breeds were used. Ten lambs were inoculated intravenously with a stabilate of an ovine E. phagocytophila strain. On the third day of fever, 4 lambs were given long-acting oxytetracycline (Terramycin prolongatum vet, Pfizer) (20 mg/kg) intramuscularly and another 4 lambs were given short-acting oxytetracycline (Terramycin vet, Pfizer) (10 mg/kg) intravenously for 5 consecutive days. The lambs were examined for the presence of Ehrlichia infection by blood smear evaluation, polymerase chain reaction (PCR) and antibody titre against E. equi. One month after the last antibiotic treatment, 250 ml citrate blood from each of these lambs were inoculated into each of 10 susceptible lambs, which were observed during the following 6 weeks. The results indicate that oxytetracycline given in the acute stage of the infection may effectively terminate the development of fever, rickettsemia and weight reduction in E. phagocytophila infected lambs. No difference was observed between the 2 treatment groups. However, at least 3 of 8 antibiotic treated lambs (37.5%) were still infected with granulocytic Ehrlichia 3 months after treatment.

The use of doxycycline in the treatment of canine ehrlichiosis.

The use of doxycycline in the treatment of twenty dogs with canine ehrlichiosis is described. The drug was found to be effective even in cases which did not respond to treatment with oxytetracycline.

Lack of susceptibility of Ehrlichia canis to imidocarb dipropionate in vitro.

In vitro antimicrobial susceptibility testing was used to compare the efficacy of imidocarb dipropionate and doxycycline on the growth of Ehrlichia canis in DH82 cell cultures. Over a 9-day period there were no significant differences (p < 0.01) in the growth of E. canis in untreated control wells and those to which imidocarb dipropionate was added at 1.2, 2.4, 4.8 or 12 micrograms/ml for the 1st 3 days. Average infection rates rose from 50 to 55% on day 0 to 100% on day 5 or 6. Doxycycline at 1 microgram/ml had residual or rickettsiocidal activity against E. canis with the average percentages of DH82 cells infected declining from 51 to 24% while the organism was exposed to the drug (3 days) and from 21 to 2% in the 6 days following removal of the drug from the cell culture medium.

Further evidence for the efficacy of imidocarb dipropionate in the treatment of Ehrlichia canis infection.

Three dogs experimentally infected with Ehrlichia canis developed thrombocytopaenia and high antibody titres to E. canis in indirect fluorescent antibody tests. One dog also became leukopaenic. At Weeks 6 and 8 post-infection, the dogs were treated with imidocarb dipropionate (5 mg kg-1 subcutaneously) and a further dose was administered at Week 12 (5 mg kg-1 intramuscularly). Twelve weeks after the last treatment (post-treatment), all dogs had normal platelet counts which persisted for a further 10 weeks until the end of the experiment. The leukopaenia resolved 20 weeks post-treatment. Although antibody titres (< 1/5) to E. canis could not be detected prior to infection, titres of 1/2 560 to 1/5 120 developed by Week 6. By Week 8 post-treatment titres began to decline and by the end of the experiment were 5- to 6-fold serum dilutions lower (1/80 to 1/320). Sub-inoculation experiments 18 weeks post-treatment, failed to cause disease or stimulate antibody responses in susceptible dogs. Serology and sub-inoculation studies on 2 dogs experimentally infected with E. canis but not treated with imidocarb dipropionate, showed that these animals remained infected for the duration of the experiment. The results of these experiments confirm that imidocarb dipropionate is effective in the treatment of canine ehrlichiosis.

In vitro antibiotic susceptibility of the newly recognized agent of ehrlichiosis in humans, Ehrlichia chaffeensis.

Ehrlichiosis in humans, a rickettsial disease recently discovered in the United States, is generally treated successfully with tetracyclines; however treatment with these agents is usually avoided with children and pregnant women. The in vitro susceptibility of Ehrlichia chaffeensis, the agent of human ehrlichiosis in the United States, was assessed by a quantitative evaluation of infected DH82 cells cultivated in 96-well microtiter plates in the presence of different concentrations of selected antibiotics. Extracellular MICs and MBCs were evaluated after 72 h of exposure to the antibiotics. Doxycycline and rifampin were found to exert rapidly bactericidal effects, with MBCs in the extracellular culture medium of less than 0.5 and 0.125 microgram/ml, respectively. E. chaffeensis was resistant to chloramphenicol, ciprofloxacin, erythromycin, co-trimoxazole, penicillin, and gentamicin, which had MICs greater than 16, 4, 8, 4, 40, and 32 micrograms/ml, respectively. These observations are consistent with the finding that human ehrlichiosis appears to respond to tetracycline therapy, which has been the therapy of first choice. Further clinical investigations are necessary to evaluate the role of rifampin in the treatment of human ehrlichiosis, especially in children.

Antibiotic susceptibility of the newly cultivated agent of human granulocytic ehrlichiosis: promising activity of quinolones and rifamycins.

Human granulocytic ehrlichiosis (HGE) is a rapidly emerging tick-borne infection which presents as an acute febrile illness and is associated with hematologic abnormalities, elevated hepatic transaminase levels, and characteristic intracellular organisms in peripheral blood granulocytes. Although HGE has been successfully treated with tetracyclines, its susceptibility to other antibiotics remains unknown. No clear treatment alternative exist for young children, pregnant women, or allergic individuals, in whom tetracyclines are contra-indicated. We performed in vitro antibiotic susceptibility tests with this recently isolated agent grown in the human promyelocytic leukemia cell line HL-60. Doxycycline (MIC, 0.25 micrograms/ml), rifampin (MIC, 0.5 micrograms/ml), rifabutin (MIC, < or = 0.125 micrograms/ml), ciprofloxacin and ofloxacin (both with MICs of 2 micrograms/ml), and trovafloxacin (MIC, < or = 0.125 micrograms/ml) ciprofloxacin and ofloxacin (both with MICs of 2 micrograms/ml), and trovafloxacin (MIC, < or = 0.125 micrograms/ml) demonstrated significant activity against the HGE agent. These agents were also bactericidal. The HGE agent was resistant to clindamycin, trimethoprim-sulfamethoxazole, and imipenem-cilastatin, as well as to ampicillin, ceftriaxone, erythromycin, and azithromycin, antibiotics commonly used to treat Lyme disease. Both chloramphenicol and gentamicin had weak inhibitory activities but were not bactericidal. Our findings confirm the observed clinical efficacy of doxycycline and further suggest that the rifamycins and quinolones, particularly trovafloxacin, hold promise as alternative agents for treating this new infection.

In vitro susceptibility of Ehrlichia sennetsu to antibiotics.

Antibiotic efficacies were evaluated by Diff-Quik (Dade, Düdingen, Federal Republic of Germany) staining of Ehrlichia sennetsu in P388D1 murine macrophages grown in 96-well microtiter plates. Sennetsu disease is generally cured with tetracyclines. In vivo, E. sennetsu is susceptible to doxycycline and is resistant to erythromycin, penicillin, and chloramphenicol. Our study confirmed, in vitro, the efficacy of doxycycline, which had an MIC of 0.125 micrograms/ml. E. sennetsu was found to be resistant to erythromycin, chloramphenicol, penicillin, gentamicin, and co-trimoxazole, while it was very susceptible to ciprofloxacin (MIC, 0.125 micrograms/ml) and rifampin (MIC, 0.5 micrograms/ml).

Antimicrobial susceptibility of Ehrlichia phagocytophila.

Human granulocytic ehrlichiosis is a recently described disease caused by an obligate intracellular gram-negative organism recently named Ehrlichia phagocytophila. To expand our knowledge of the susceptibility of E. phagocytophila, we tested six New York State isolates for susceptibility to 12 antimicrobials using an HL-60 cell culture system. All of the isolates were susceptible to doxycycline (MIC, < or =0.125 microg/ml; minimum bactericidal concentration [MBC], 0.125 to 0.5 microg/ml), rifampin (MIC, < or =0.125 microg/ml; MBC, < or =0.125 microg/ml), ofloxacin (MIC, < or =2 microg/ml; MBC, < or =2 microg/ml), levofloxacin (MIC, < or =1 microg/ml; MBC, < or =1 microg/ml), and trovafloxacin (MIC, < or =0.032 microg/ml; MBC, < or =0.032 microg/ml). Isolates were uniformly resistant to amoxicillin, ceftriaxone, erythromycin, azithromycin, clarithromycin, and amikacin. For one strain, the MBC of chloramphenicol was < or =8 microg/ml. These data suggest that quinolone antibiotics and rifampin may be alternative agents for patients with intolerance to tetracyclines.

Doxycycline hyclate treatment of experimental canine ehrlichiosis followed by challenge inoculation with two Ehrlichia canis strains.

Dogs were experimentally inoculated with Ehrlichia canis Florida to assess the efficacy of doxycycline hyclate for the treatment of acute ehrlichiosis. Treatment with doxycycline eliminated infection in eight of eight dogs. Untreated infected control dogs appeared to eliminate the infection or, alternatively, suppress the degree of ehrlichiemia to a level not detectable by tissue culture isolation or PCR or by transfusion of blood into recipient dogs. Prior infection did not infer protection against homologous (strain Florida) or heterologous (strain NCSU Jake) strains of E. canis. We conclude that doxycycline hyclate is an effective treatment for acute E. canis infection; however, these results may not be applicable to chronic infections in nature. Spontaneous resolution of infection, induced by the dog's innate immune response, provides evidence that an E. canis vaccine, once developed, might potentially confer protective immunity against the organism.

Role of Ca2+ and calmodulin in ehrlichial infection in macrophages.

Replication of Ehrlichia risticii was inhibited in P388D1 cells and murine peritoneal macrophages when a calmodulin antagonist (W-7, chlorpromazine, or trifluoperazine); a Ca2+ channel blocker (verapamil, diltiazem, nifedipine, or flunarizine); an extracellular Ca2+ chelator, EGTA [ethylene glycol-bis(beta-aminoethyl ether)-N,N,N',N'-tetraacetic acid]; an inhibitor of intracellular Ca2+ mobilization, TMB-8; or Ca2+ ionophore A23187 was added after internalization of the organism at 3 h postincubation. When intracellular ehrlichiae at their logarithmic stage of growth were treated with these reagents, not only was further proliferation prevented but also there was significant reduction in numbers of intracellular ehrlichiae. These reagents prevented spreading of E. risticii from P388D1 cells to THP-1 cells. None of these reagents prevented binding of [35S]methionine-labeled E. risticii to P388D1 cells, but all of these reagents prevented internalization of [35S]methionine-labeled E. risticii. Protein kinase C inhibitors, H-7 and staurosporin, had no effect. 14CO2 production from L-[14C]glutamine in Percoll-density-gradient-purified E. risticii was inhibited by A23187 but not by W-7 or verapamil, suggesting that Ca2+ but not calmodulin directly regulates ehrlichials glutamine oxidation. Pretreatment of E. risticii with W-7 or verapamil did not reduce its infectivity. These results indicate that calmodulin and Ca2+ are essential for ehrlichial internalization, replication, and spreading in macrophages but are not essential for binding.

L-arginine-dependent killing of intracellular Ehrlichia risticii by macrophages treated with gamma interferon.

Thioglycolate-induced murine peritoneal macrophages infected with Ehrlichia risticii and treated in vitro with gamma interferon (IFN-gamma) developed antiehrlichial activity that eliminated the intracellular bacteria. This antiehrlichial activity was suppressed by NG-monomethyl-L-arginine, a competitive inhibitor of nitric oxide synthesis from L-arginine, but not by L-tryptophan. Increased levels of nitrite, an oxidative product of nitric oxide, were measured in cultures of infected macrophages treated with IFN-gamma. Sodium nitroprusside, which spontaneously releases nitric oxide, also showed the antiehrlichial activity. The antiehrlichial activity by reactive nitrogen intermediates was not mediated by elevation of the cellular concentration of cyclic GMP since the addition of 8-bromo-cyclic GMP itself had no influence on ehrlichial infection of macrophages. Addition of the intracellular iron chelator deferoxamine also inhibited E. risticii infection in vitro. These results suggest that intracellular E. risticii survival is iron dependent and that production of reactive nitrogen intermediates triggers iron loss from critical target enzymes of E. risticii, leading to lethal metabolic inhibition. However, addition of excess FeSO4, ferric citrate, or iron-saturated transferrin did not counteract the antiehrlichial effect induced by IFN-gamma.

Tyrosine phosphorylation is required for ehrlichial internalization and replication in P388D1 cells.

Replication of Ehrlichia risticii was inhibited in P388D1 cells when a protein tyrosine kinase inhibitor (genistein or herbimycin A) was added after internalization of the organism at 3 h postinfection. Upon addition of genistein at day 1, 2, 3, or 4 postinfection, further proliferation of E. risticii was prevented. The inhibition was reversible, since regrowth of E. risticii occurred upon the removal of genistein. Genistein prevented spreading of E. risticii from P388D1 cells to THP-1 cells. Genistein did not prevent binding of [35S]methionine-labeled E. risticii to P388D1 cells but did prevent internalization of [35S]methionine-labeled E. risticii. 14CO2 production from L-[14C]glutamine in Percoll density gradient-purified E. risticii was not inhibited by genistein or herbimycin A, which suggests that these reagents did not directly inhibit ehrlichial energy metabolism. Double indirect immunofluorescence labeling with antiphosphotyrosine antibody and anti-E. risticii antibody revealed colocalization of tyrosine phosphoproteins with ehrlichial inclusions. There was, however, no colocalization of phosphotyrosine with phagosomes containing 0.5-microm-diameter fluorescent beads. Western immunoblot analysis revealed that 52- and 54-kDa proteins were tyrosine phosphorylated only in infected cells and that phosphorylation of these two proteins was reduced when infected cells were treated with genistein for 6 h. These results suggest that protein tyrosine phosphorylation is specific and essential for ehrlichial internalization, replication, and spreading in macrophages but not for binding.

DNA gyrase-mediated natural resistance to fluoroquinolones in Ehrlichia spp.

Fluoroquinolone susceptibility heterogeneity between various Ehrlichia species has been previously demonstrated. In gram-negative bacteria, resistance to fluoroquinolones most often corresponds to specific amino acid variations in a portion of the protein sequence of the A subunit of DNA gyrase (GyrA), referred to as the quinolone resistance-determining region (QRDR). We suspected a similar mechanism to be responsible for natural resistance in some Ehrlichia species. To verify this hypothesis, we sequenced the entire gyrA gene of the quinolone-susceptible species Ehrlichia sennetsu and designed specific primers to amplify and sequence the QRDR of four other Ehrlichia species as well as the closely related species Cowdria ruminantium. We identified in the fluoroquinolone-resistant species Ehrlichia chaffeensis and Ehrlichia canis a specific GyrA QRDR amino acid sequence, also present in C. ruminantium (whose susceptibility to fluoroquinolones remains unknown). These three species belong to a single phylogenetic cluster referred to as the E. canis genogroup. A different GyrA QRDR pattern, shared by the Ehrlichia species representatives of the E. sennetsu and Ehrlichia phagocytophila genogroups, was identified. Three of the four species tested are known to be susceptible to fluoroquinolones. A serine residue in position 83 (Escherichia coli numbering) in the susceptible species is replaced by an alanine residue in fluoroquinolone-resistant species. These results are consistent with the current knowledge on fluoroquinolone resistance in other gram-negative bacteria. They are indicative of a natural gyrase-mediated resistance to fluoroquinolones in the E. canis genogroup.

In vitro susceptibilities of Ehrlichia risticii to eight antibiotics.

Inhibition of the proliferation of Ehrlichia risticii cultured in murine macrophage P388D1 cells by eight antibiotics was evaluated by indirect fluorescent-antibody staining with an antiserum specific to E. risticii. There was a negative correlation between the percentage of infected cells and the log10 of the concentrations of all antibiotics examined. The ranks of the antibiotics in the order of 50% inhibitory concentrations (on a microgram-per-milliliter basis) after 48 h of exposure were as follows: demeclocycline, doxycycline, and oxytetracycline less than minocycline less than rifampin less than tetracycline less than erythromycin and nalidixic acid. When the antibiotics were removed after 48 h of incubation, continuous inhibition of proliferation was evident at 72 h. At 96 h regrowth of the organisms occurred in most of the cultures. The rate of regrowth was the highest with nalidixic acid, followed by erythromycin, at all concentrations of the antibiotic tested. Regrowth was observed with less than 0.1 microgram of minocycline per ml and less than 0.01 microgram of oxytetracycline, tetracycline, and doxycycline per ml. With more than 0.01 microgram of demeclocycline per ml, however, the inhibition persisted for up to 72 h after removal of the antibiotic. These results indicate that demeclocycline was slightly more effective than doxycycline, oxytetracycline, and minocycline in eliminating E. risticii in macrophages in vitro, whereas tetracycline and rifampin were less effective. Nalidixic acid and erythromycin were ineffective.